code/15.Examine-GA-ICELLNET.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import cell2cell as c2c\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "\n",
    "import glob\n",
    "import json\n",
    "from collections import defaultdict\n",
    "\n",
    "import scipy\n",
    "import scipy.cluster.hierarchy as hc\n",
    "\n",
    "from sklearn.metrics import pairwise_distances\n",
    "\n",
    "from cci_distance_correlation import get_correlation\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analysis Setup\n",
    "\n",
    "Here, all inputs and parameters needed for the analysis are defined"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Location of files**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_folder = '../data/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "output_folder = '../results/CCI-GA-ICELLNET-LRs/'\n",
    "if not os.path.isdir(output_folder):\n",
    "    os.mkdir(output_folder)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "files = dict()\n",
    "files['rnaseq'] = data_folder + '/RNA-Seq/Celegans_RNASeqData_Cell.xlsx'\n",
    "files['metadata'] = data_folder + '/RNA-Seq/Celegans_cell_metadata.tsv'\n",
    "files['ppi'] = data_folder + '/PPI-Networks/Celegans-Curated-LR-pairs.xlsx'\n",
    "files['phenotype'] = data_folder + '/Digital-3D-Map/Celegans_Physical_Distances_Min.csv'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**RNA-seq data manipulation**\n",
    "\n",
    "- <font color=\"green\">'gene_col'</font> is the name of the column containing the list or names of genes in the dataset.\n",
    "- <font color=\"green\">'drop_nangenes'</font> is a boolean decision to remove genes with NaN values.\n",
    "- <font color=\"green\">'log_transform'</font> is a boolean decision to log-transform the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "rnaseq_setup = dict()\n",
    "rnaseq_setup['gene_col'] = 'gene_id'\n",
    "rnaseq_setup['drop_nangenes'] = True\n",
    "rnaseq_setup['log_transform'] = False"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Metadata columns**\n",
    "\n",
    "- <font color=\"green\">'sample_col'</font> is the name of the column containing the sample or cell names in RNA-seq data (equivalent to the columns in rnaseq_data.\n",
    "- <font color=\"green\">'group_col'</font> is the name of the column containing the labels for the respective samples or cells in order to group/color them in downstream analyses."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "meta_setup = dict()\n",
    "meta_setup['sample_col'] = '#SampleID'\n",
    "meta_setup['group_col'] = 'Groups'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Columns for PPI network**\n",
    "\n",
    "- <font color=\"green\">'protein_cols'</font> is a list containing the names of the first and second interactors."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "ppi_setup = dict()\n",
    "ppi_setup['protein_cols'] = ['Ligand_WB', 'Receptor_WB']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Cutoffs setup for binarizing gene expression**\n",
    "\n",
    "- <font color=\"green\">'type'</font> is the type of cutoff to use. \n",
    "    - 'constant_value'\n",
    "    - 'local_percentile'\n",
    "    - 'global_percentile'\n",
    "    - None\n",
    "- <font color=\"green\">'parameter'</font> is the corresponding parameter for the type of cutoff. If percentile is used, it should be a value between 0 and 1. E.g. a value of 0.75 represents the 75-th percentile, and the cutoff to binarize the gene expression will be that value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "cutoff_setup = dict()\n",
    "cutoff_setup['type'] = 'constant_value'\n",
    "cutoff_setup['parameter'] = 10 # TPM"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**General Analysis setups**\n",
    "\n",
    "- <font color=\"green\">'communication_score'</font> is the type of communication score used to detect active pathways between each pair of cell. It can be:\n",
    "    - 'expression_thresholding'\n",
    "    - 'expression_product'\n",
    "    - 'expression_correlation'\n",
    "    - 'differential_combinations'\n",
    "- <font color=\"green\">'cci_score'</font> is the scoring function to aggregate the communication scores. It can be:\n",
    "    - 'bray_curtis'\n",
    "    - 'jaccard'\n",
    "    - 'count'\n",
    "- <font color=\"green\">'cci_type'</font> is the type of interaction between two cells. If it is undirected, ligand and receptor are considered from both cells. If it is directed, ligands from one cell and receptors from the other are considered separately with respect to ligands from the second cell and receptor from the first one. So, it can be:\n",
    "    - 'undirected'\n",
    "    - 'directed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "analysis_setup = dict()\n",
    "analysis_setup['communication_score'] = 'expression_product'\n",
    "analysis_setup['cci_score'] = 'icellnet'\n",
    "analysis_setup['cci_type'] = 'undirected'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**RNA-seq**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Opening RNAseq datasets from ../data//RNA-Seq/Celegans_RNASeqData_Cell.xlsx\n",
      "../data//RNA-Seq/Celegans_RNASeqData_Cell.xlsx was correctly loaded\n"
     ]
    }
   ],
   "source": [
    "rnaseq_data = c2c.io.load_rnaseq(rnaseq_file=files['rnaseq'],\n",
    "                                 gene_column=rnaseq_setup['gene_col'],\n",
    "                                 drop_nangenes=rnaseq_setup['drop_nangenes'],\n",
    "                                 log_transformation=rnaseq_setup['log_transform'],\n",
    "                                 format='auto')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "rnaseq_data = rnaseq_data.applymap(lambda x: np.log2(x+1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Metadata for cell types**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "../data//RNA-Seq/Celegans_cell_metadata.tsv was correctly loaded\n"
     ]
    }
   ],
   "source": [
    "meta = c2c.io.load_metadata(metadata_file=files['metadata'],\n",
    "                            cell_labels=rnaseq_data.columns,\n",
    "                            format='auto')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**PPI network**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Opening PPI datasets from ../data//PPI-Networks/Celegans-Curated-LR-pairs.xlsx\n",
      "../data//PPI-Networks/Celegans-Curated-LR-pairs.xlsx was correctly loaded\n",
      "Removing bidirectionality of PPI network\n",
      "Simplying PPI network\n"
     ]
    }
   ],
   "source": [
    "ppi_data = c2c.io.load_ppi(ppi_file=files['ppi'],\n",
    "                           interaction_columns=ppi_setup['protein_cols'],\n",
    "                           rnaseq_genes=list(rnaseq_data.index),\n",
    "                           format='auto')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(245, 3)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ppi_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "symbol_lrs = pd.read_excel(files['ppi'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Ligand_WB</th>\n",
       "      <th>Receptor_WB</th>\n",
       "      <th>Ligand_symbol</th>\n",
       "      <th>Receptor_symbol</th>\n",
       "      <th>LR Function</th>\n",
       "      <th>L Function</th>\n",
       "      <th>R Function</th>\n",
       "      <th>Ligand_desc_WormBase_versionWS277</th>\n",
       "      <th>Receptor_desc_WormBase_versionWS277</th>\n",
       "      <th>Pair_confidence</th>\n",
       "      <th>Location_type</th>\n",
       "      <th>Confidence_evidence</th>\n",
       "      <th>Location_evidence</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>WBGene00000182</td>\n",
       "      <td>WBGene00004040</td>\n",
       "      <td>arf-1.2</td>\n",
       "      <td>pld-1</td>\n",
       "      <td>Cell adhesion</td>\n",
       "      <td>ADP-ribosylation factor 1</td>\n",
       "      <td>Phospholipase D2</td>\n",
       "      <td>Is an ortholog of human ARF1 (ADP ribosylation...</td>\n",
       "      <td>Is an ortholog of human PLD2 (phospholipase D2...</td>\n",
       "      <td>Putative based on data</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase: arf-1 found in the endoplasmic retic...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>WBGene00000903</td>\n",
       "      <td>WBGene00004860</td>\n",
       "      <td>daf-7</td>\n",
       "      <td>sma-6</td>\n",
       "      <td>TGF-B signaling</td>\n",
       "      <td>GDF11</td>\n",
       "      <td>BMPRIB</td>\n",
       "      <td>Is an ortholog of human GDF11 (growth differen...</td>\n",
       "      <td>Is an ortholog of human BMPR1A (bone morphogen...</td>\n",
       "      <td>Putative based on literature</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase; \" Located in extracellular space,\" h...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>WBGene00005647</td>\n",
       "      <td>WBGene00008779</td>\n",
       "      <td>srp-6</td>\n",
       "      <td>F14B4.1</td>\n",
       "      <td>Wnt signaling</td>\n",
       "      <td>Serpin family (Blocks wnt signaling)</td>\n",
       "      <td>Wnt-Receptor</td>\n",
       "      <td>Is an ortholog of human SERPINB10 (serpin fami...</td>\n",
       "      <td>Is predicted to have Wnt-activated receptor ac...</td>\n",
       "      <td>Putative based on literature</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase: srp-6 is found and active in the ext...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>WBGene00000074</td>\n",
       "      <td>WBGene00003930</td>\n",
       "      <td>adm-2</td>\n",
       "      <td>pat-3</td>\n",
       "      <td>Cell adhesion</td>\n",
       "      <td>ADAM</td>\n",
       "      <td>Integrin</td>\n",
       "      <td>Is an ortholog of several human genes includin...</td>\n",
       "      <td>Is an ortholog of human ITGB1 (integrin subuni...</td>\n",
       "      <td>Putative based on data</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase: adm-2 is an integral component of th...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>WBGene00000074</td>\n",
       "      <td>WBGene00003929</td>\n",
       "      <td>adm-2</td>\n",
       "      <td>pat-2</td>\n",
       "      <td>Cell adhesion</td>\n",
       "      <td>ADAM</td>\n",
       "      <td>Integrin</td>\n",
       "      <td>Is an ortholog of several human genes includin...</td>\n",
       "      <td>Is an ortholog of human ITGA5 (integrin subuni...</td>\n",
       "      <td>Putative based on data</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase: adm-2 is an integral component of th...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Ligand_WB     Receptor_WB Ligand_symbol Receptor_symbol  \\\n",
       "0  WBGene00000182  WBGene00004040       arf-1.2           pld-1   \n",
       "1  WBGene00000903  WBGene00004860         daf-7           sma-6   \n",
       "2  WBGene00005647  WBGene00008779         srp-6         F14B4.1   \n",
       "3  WBGene00000074  WBGene00003930         adm-2           pat-3   \n",
       "4  WBGene00000074  WBGene00003929         adm-2           pat-2   \n",
       "\n",
       "       LR Function                            L Function        R Function  \\\n",
       "0    Cell adhesion             ADP-ribosylation factor 1  Phospholipase D2   \n",
       "1  TGF-B signaling                                 GDF11            BMPRIB   \n",
       "2    Wnt signaling  Serpin family (Blocks wnt signaling)      Wnt-Receptor   \n",
       "3    Cell adhesion                                  ADAM          Integrin   \n",
       "4    Cell adhesion                                  ADAM          Integrin   \n",
       "\n",
       "                   Ligand_desc_WormBase_versionWS277  \\\n",
       "0  Is an ortholog of human ARF1 (ADP ribosylation...   \n",
       "1  Is an ortholog of human GDF11 (growth differen...   \n",
       "2  Is an ortholog of human SERPINB10 (serpin fami...   \n",
       "3  Is an ortholog of several human genes includin...   \n",
       "4  Is an ortholog of several human genes includin...   \n",
       "\n",
       "                 Receptor_desc_WormBase_versionWS277  \\\n",
       "0  Is an ortholog of human PLD2 (phospholipase D2...   \n",
       "1  Is an ortholog of human BMPR1A (bone morphogen...   \n",
       "2  Is predicted to have Wnt-activated receptor ac...   \n",
       "3  Is an ortholog of human ITGB1 (integrin subuni...   \n",
       "4  Is an ortholog of human ITGA5 (integrin subuni...   \n",
       "\n",
       "                Pair_confidence   Location_type  \\\n",
       "0        Putative based on data  Membrane-bound   \n",
       "1  Putative based on literature        Secreted   \n",
       "2  Putative based on literature        Secreted   \n",
       "3        Putative based on data  Membrane-bound   \n",
       "4        Putative based on data  Membrane-bound   \n",
       "\n",
       "                                 Confidence_evidence  \\\n",
       "0  Orthology, protein-protein interaction (STRING...   \n",
       "1  Orthology, protein-protein interaction (STRING...   \n",
       "2  Orthology, protein-protein interaction (STRING...   \n",
       "3  Orthology, protein-protein interaction (STRING...   \n",
       "4  Orthology, protein-protein interaction (STRING...   \n",
       "\n",
       "                                   Location_evidence  \n",
       "0  wormbase: arf-1 found in the endoplasmic retic...  \n",
       "1  wormbase; \" Located in extracellular space,\" h...  \n",
       "2  wormbase: srp-6 is found and active in the ext...  \n",
       "3  wormbase: adm-2 is an integral component of th...  \n",
       "4  wormbase: adm-2 is an integral component of th...  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "symbol_lrs.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "tmp = ppi_data.copy()\n",
    "tmp.columns = ['Ligand_WB', 'Receptor_WB', 'score']\n",
    "symbol_lrs = tmp.merge(symbol_lrs, how='inner').drop_duplicates().reset_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Phenotype**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "physical_distance = pd.read_csv(files['phenotype'], index_col=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cells to include"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "included_cells = sorted(list(set(rnaseq_data.columns) & set(physical_distance.columns)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "excluded_cells = sorted(list(set(rnaseq_data.columns) - set(included_cells)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Distal_tip_cells',\n",
       " 'Sex_myoblasts',\n",
       " 'Socket_cells',\n",
       " 'Vulval_precursors',\n",
       " 'flp-1(+)_interneurons']"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "excluded_cells"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Collect & Preprocess GA Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "../data//GA-ICELLNET/0141aa97-893c-4d25-9d69-5a1cde3c36b6.json\n",
      "../data//GA-ICELLNET/043c0c44-cb36-4933-aa45-9ba70e004137.json\n",
      "../data//GA-ICELLNET/04eeaf98-7f8d-4b90-afcc-a4f9f2294777.json\n",
      "../data//GA-ICELLNET/07ec168f-f388-4b7e-a49a-604786ee3f3c.json\n",
      "../data//GA-ICELLNET/09b258c9-fdfc-494e-b9e5-64db43d0fe1b.json\n",
      "../data//GA-ICELLNET/0cfdcee9-0f8e-4358-9c3a-0a796ed09399.json\n",
      "../data//GA-ICELLNET/0e6b3adc-9f2b-4dcd-a22e-6674784b5c72.json\n",
      "../data//GA-ICELLNET/0ede2cc0-45d3-4766-8a45-7d54a3e59f71.json\n",
      "../data//GA-ICELLNET/100ca935-c1eb-4b07-a726-84a0c19291d4.json\n",
      "../data//GA-ICELLNET/125c57d5-52e8-47e9-aea0-b5b2682cc551.json\n",
      "../data//GA-ICELLNET/189658b0-cf4d-4192-ad93-99c4ca5d551c.json\n",
      "../data//GA-ICELLNET/1a21bc2b-5e9c-4c4b-b482-01cef524b8e2.json\n",
      "../data//GA-ICELLNET/1b3a7d48-b2b6-4da1-bbd3-cf2658b7129a.json\n",
      "../data//GA-ICELLNET/1b60b00f-fe09-4777-95be-70f4d42281a2.json\n",
      "../data//GA-ICELLNET/1c428010-feb4-499d-8fce-4ec6e5dea6cb.json\n",
      "../data//GA-ICELLNET/1f6be7f3-70ae-4f47-b4fa-409f79b3ebfe.json\n",
      "../data//GA-ICELLNET/252b1ebf-0624-47d8-95d5-c28f9cd77e03.json\n",
      "../data//GA-ICELLNET/276be70c-fc7f-43fd-a65c-b7b0f38b4ddb.json\n",
      "../data//GA-ICELLNET/2875150e-7312-457a-92cd-d6a84fec616d.json\n",
      "../data//GA-ICELLNET/2a62150b-ad03-408f-8392-b9bdd6305121.json\n",
      "../data//GA-ICELLNET/2ab5cf8e-ebd2-43bc-b0e5-18bd6390faaa.json\n",
      "../data//GA-ICELLNET/2f2572d4-b9fa-4e84-93bc-87b6261d58f1.json\n",
      "../data//GA-ICELLNET/348aecab-209d-475a-8df3-61fa4a16925d.json\n",
      "../data//GA-ICELLNET/350f73fe-d489-4a95-80bf-86c9d6fe4bc1.json\n",
      "../data//GA-ICELLNET/369b0e24-031a-4036-8848-1ee56babcaf9.json\n",
      "../data//GA-ICELLNET/36f13b48-5453-401a-9dfb-34d5bb62f7d4.json\n",
      "../data//GA-ICELLNET/3cab52a6-1450-45c1-b905-b1730da6e2e2.json\n",
      "../data//GA-ICELLNET/3dda82ca-092c-460b-81fd-42f982751ad1.json\n",
      "../data//GA-ICELLNET/40989ce3-b1b9-478b-9a44-fb9322d89a7e.json\n",
      "../data//GA-ICELLNET/49a9f8a6-ab6b-4903-9bfc-0f3bc447a309.json\n",
      "../data//GA-ICELLNET/4b0ce76a-bc96-4ba7-be64-a4e49b1c4bd3.json\n",
      "../data//GA-ICELLNET/4b552d5f-937f-411d-8fd9-716588cb9c2b.json\n",
      "../data//GA-ICELLNET/51a7c32f-27cb-41b8-ae19-6b0231797203.json\n",
      "../data//GA-ICELLNET/530a9957-ff2d-4b94-9ab6-3500d05a4007.json\n",
      "../data//GA-ICELLNET/55e1dc91-8d02-4acd-8af0-3b2d955ec262.json\n",
      "../data//GA-ICELLNET/588ea07b-579d-4aec-8d51-5f8f720474c9.json\n",
      "../data//GA-ICELLNET/5e3346fa-656c-4cb0-9bea-f2479c800848.json\n",
      "../data//GA-ICELLNET/5e522baf-afbd-4232-9f77-0f1e6184f618.json\n",
      "../data//GA-ICELLNET/64c71e7d-cac6-4e46-81da-f55b59996a56.json\n",
      "../data//GA-ICELLNET/653585ca-7dde-439e-beea-0202d47b2c9d.json\n",
      "../data//GA-ICELLNET/693baa61-0e9c-4161-964f-90b94486987d.json\n",
      "../data//GA-ICELLNET/6cc43d30-e387-46d9-8754-32252d95a3fa.json\n",
      "../data//GA-ICELLNET/701308ad-a9cc-4e20-ba35-ecf1eeaef3e3.json\n",
      "../data//GA-ICELLNET/72b8387a-fe2f-491b-b3c3-1f0f273ff6e0.json\n",
      "../data//GA-ICELLNET/733b9245-2181-4496-9428-f72680930289.json\n",
      "../data//GA-ICELLNET/73534bdb-1f33-488b-b439-35adba88aed2.json\n",
      "../data//GA-ICELLNET/74c26a41-9d19-44c1-a238-c107af6c6c20.json\n",
      "../data//GA-ICELLNET/7cbf9992-c0bd-40d3-9dd1-f11c3c6db30e.json\n",
      "../data//GA-ICELLNET/7dd174a4-2bdb-41da-87ae-565b49bc15c6.json\n",
      "../data//GA-ICELLNET/7ffb15ef-9642-4075-bc8e-cd2f115d2b50.json\n",
      "../data//GA-ICELLNET/8379b93d-f752-48d3-9e9e-9e9412a06c6f.json\n",
      "../data//GA-ICELLNET/848ba72b-5a0f-4269-b48d-9a0e75fb59d8.json\n",
      "../data//GA-ICELLNET/886e9d68-8a6f-4427-ba8b-6dfd118f5dbc.json\n",
      "../data//GA-ICELLNET/8f528261-db8a-499a-94dd-50bed15d37e7.json\n",
      "../data//GA-ICELLNET/90368a49-0bd3-43f0-99b4-d042b408e5cb.json\n",
      "../data//GA-ICELLNET/9a306d9c-7621-476a-8b2b-f457632dd8fe.json\n",
      "../data//GA-ICELLNET/9c21695e-dda4-4f62-9642-fd25db6c7e2d.json\n",
      "../data//GA-ICELLNET/9e6d8c52-801e-4d8c-bb07-66db3d89438e.json\n",
      "../data//GA-ICELLNET/a0049d68-0cc5-45ed-831c-c8cd9fca860d.json\n",
      "../data//GA-ICELLNET/a6efb6a1-96d4-4587-9fa7-83a6c410bb3e.json\n",
      "../data//GA-ICELLNET/ab660410-6d26-4f66-9ba0-07b001a338c3.json\n",
      "../data//GA-ICELLNET/ac5509d5-7563-431e-a8d5-e3dd815173d8.json\n",
      "../data//GA-ICELLNET/adeb8127-6a70-404b-b2ed-f0c17b713f0e.json\n",
      "../data//GA-ICELLNET/aeba9783-37ff-461f-835e-781140b58a1d.json\n",
      "../data//GA-ICELLNET/b1a7cbe2-4cb3-42ec-babb-b1a6ea09a807.json\n",
      "../data//GA-ICELLNET/b2bf84d9-8f4b-41ea-9b29-9359a7be8f8f.json\n",
      "../data//GA-ICELLNET/b42035eb-941e-4b5a-86a8-d22bbfcc59bd.json\n",
      "../data//GA-ICELLNET/b5abecfb-6590-42bc-9577-28a261e568c8.json\n",
      "../data//GA-ICELLNET/b81802a7-43b0-45aa-b664-1351f0059d44.json\n",
      "../data//GA-ICELLNET/ba75d0d5-a074-4edd-9369-74ee50950dd4.json\n",
      "../data//GA-ICELLNET/baa3e334-5136-4c01-a445-7c12a95b8493.json\n",
      "../data//GA-ICELLNET/c02d6346-c991-40de-96f0-5a56ee6d4011.json\n",
      "../data//GA-ICELLNET/c145dd97-d885-4a07-b1a0-e2b8b335ca36.json\n",
      "../data//GA-ICELLNET/c2e462b3-5163-4d14-b152-0b9b8fe416bc.json\n",
      "../data//GA-ICELLNET/c4bdedfb-a63b-408e-803a-49710e685b44.json\n",
      "../data//GA-ICELLNET/c5475c8e-2ee0-419b-bdc2-01ddc6feed44.json\n",
      "../data//GA-ICELLNET/cb4b4ea2-edf7-4d09-96fb-f7a24da19126.json\n",
      "../data//GA-ICELLNET/cdcce419-2e51-4266-9ee8-45bc28c37a6b.json\n",
      "../data//GA-ICELLNET/d0d09e1d-ca04-4ecd-a14a-341cdbd178cf.json\n",
      "../data//GA-ICELLNET/d218f12c-358e-4d90-b997-98ce378a16b6.json\n",
      "../data//GA-ICELLNET/d65d8f76-5a76-4106-8faa-0db95f814323.json\n",
      "../data//GA-ICELLNET/d6e0bdc4-7be5-423b-8fd7-b2e6a2c49ebd.json\n",
      "../data//GA-ICELLNET/d912a7e2-a4a9-458b-8065-baf6aa088339.json\n",
      "../data//GA-ICELLNET/dab1b09c-2e7e-4587-a1f6-6428d04a2ecf.json\n",
      "../data//GA-ICELLNET/db542722-e200-4e91-b11d-cc41d32fe579.json\n",
      "../data//GA-ICELLNET/dd883e2c-fe27-43fe-ae29-621ea36d81be.json\n",
      "../data//GA-ICELLNET/dfb7125d-6636-4e58-846e-f66c4353ac1c.json\n",
      "../data//GA-ICELLNET/dfcb09b0-f49c-4781-a9b1-692bfc0ed359.json\n",
      "../data//GA-ICELLNET/e50d36a4-7370-4868-b321-85acfe1fee87.json\n",
      "../data//GA-ICELLNET/e67afdcf-a6b0-419b-a9e5-1fcc690db81f.json\n",
      "../data//GA-ICELLNET/e7bd7fa3-430a-4e93-82a9-9768974b6286.json\n",
      "../data//GA-ICELLNET/ea129b3e-0dbb-4d6f-a082-0c1325d51f8c.json\n",
      "../data//GA-ICELLNET/eb3474ce-f961-47f4-9ca7-2a350c6aaec5.json\n",
      "../data//GA-ICELLNET/ed08dc6f-855c-4721-b47e-105c96db600a.json\n",
      "../data//GA-ICELLNET/eeae80ef-103d-4360-8021-20b8fddfae11.json\n",
      "../data//GA-ICELLNET/ef4da940-fcdf-445e-b9fb-d1a8e65f7d7d.json\n",
      "../data//GA-ICELLNET/f200fbe1-2bb7-4a1c-ab95-c063602863f9.json\n",
      "../data//GA-ICELLNET/f85a0bf7-4514-4530-9592-f4faef6f652b.json\n",
      "../data//GA-ICELLNET/f87d50b0-cb65-4843-bd43-672921abca82.json\n",
      "../data//GA-ICELLNET/fb5760d7-2474-413d-ac8a-f23d56631685.json\n"
     ]
    }
   ],
   "source": [
    "all_files = sorted(glob.glob(data_folder + \"/GA-ICELLNET/*.json\"))\n",
    "\n",
    "selected_ppis = []\n",
    "score_per_run = defaultdict(list)\n",
    "\n",
    "max_corr = 0.0\n",
    "max_array = []\n",
    "for filename in all_files:\n",
    "    print(filename)\n",
    "    with open(filename) as json_file:\n",
    "        data = json.load(json_file)\n",
    "        runs = sorted([int(r.lstrip('run')) for r in data.keys()])\n",
    "        selected_ppis.append(data['run{}'.format(runs[-1])]['ppi_data'])\n",
    "        for r in runs:\n",
    "            corr = data['run{}'.format(r)]['obj_fn']\n",
    "            score_per_run[r].append(corr)\n",
    "            if corr > max_corr:\n",
    "                max_corr = corr\n",
    "                max_array = data['run{}'.format(r)]['ppi_data']\n",
    "            \n",
    "#results['run{}'.format(runs_)] = {'obj_fn' : opt_score,\n",
    "#                                          'ppi_data' : tmp_ppi['score'].values.tolist(),\n",
    "#                                          'drop_fraction' : drop_fraction\n",
    "#                                         }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(filename) as json_file:\n",
    "    data = json.load(json_file)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Ratio of selection for each LR pair**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "ratio_selected_ppis = np.nansum(np.array(selected_ppis), axis=0) / len(selected_ppis)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "ga_selected_ppi_data = ppi_data.copy()\n",
    "ga_selected_ppi_data['score'] = ratio_selected_ppis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**LR pairs selected in the list that led to maximal correlation**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "max_ga_selected = ppi_data.copy()\n",
    "max_ga_selected['score'] = np.array(max_array).astype(int)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plot results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Correlations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Correlation for basal PPI list**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "ref_distance_vector = scipy.spatial.distance.squareform(physical_distance.loc[included_cells, included_cells].values)\n",
    "\n",
    "bi_ppi_data = c2c.preprocessing.bidirectional_ppi_for_cci(ppi_data=ppi_data, verbose=False)\n",
    "\n",
    "interaction_space = c2c.core.InteractionSpace(rnaseq_data=rnaseq_data[included_cells],\n",
    "                                              ppi_data=bi_ppi_data,\n",
    "                                              gene_cutoffs=cutoff_setup,\n",
    "                                              communication_score=analysis_setup['communication_score'],\n",
    "                                              cci_score=analysis_setup['cci_score'],\n",
    "                                              cci_type=analysis_setup['cci_type'],\n",
    "                                              verbose=False)\n",
    "\n",
    "# Compute interactions\n",
    "interaction_space.compute_pairwise_cci_scores(use_ppi_score=True, verbose=False)\n",
    "\n",
    "# Distance matrix from analysis\n",
    "result_matrix = interaction_space.distance_matrix.loc[included_cells, included_cells]\n",
    "\n",
    "# Compute correlations\n",
    "result_distance_vector = scipy.spatial.distance.squareform(result_matrix)\n",
    "corr = scipy.stats.spearmanr(result_distance_vector, ref_distance_vector)\n",
    "base_result = abs(np.nan_to_num(corr[0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Structure data for visualization**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "mean_values = [base_result]\n",
    "std_values = [0.0]\n",
    "\n",
    "for k in sorted(list(score_per_run.keys())):\n",
    "    v = score_per_run[k]\n",
    "    mean_values.append(np.nanmean(v))\n",
    "    std_values.append(np.nanstd(v))\n",
    "    \n",
    "run_values = list(range(len(mean_values)))\n",
    "\n",
    "data = np.array(list(zip(run_values, mean_values, std_values)))\n",
    "df = pd.DataFrame(data, columns=['X', 'MEAN', 'STD'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Plot results**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7054949986833712"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "max_corr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.30456263973340303,\n",
       " 0.557485595263601,\n",
       " 0.6564809996695611,\n",
       " 0.6662653265569273,\n",
       " 0.6821437882908472]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.0,\n",
       " 0.005706146469299825,\n",
       " 0.00894543945249132,\n",
       " 0.011730612226509077,\n",
       " 0.009518013739417266]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "std_values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with sns.axes_style(\"darkgrid\"):\n",
    "    fig=plt.plot('X', 'MEAN', data=df, linestyle='', marker='o', markersize=8, color='b', label='Mean')\n",
    "    \n",
    "    mean_precision = np.array(mean_values)\n",
    "    std_precision = np.array(std_values)\n",
    "    precision_upper = np.minimum(mean_precision + std_precision, 1)\n",
    "    precision_lower = np.maximum(mean_precision - std_precision, 0)\n",
    "\n",
    "    plt.fill_between(run_values, precision_lower, precision_upper, color='steelblue', alpha=.2, \n",
    "                     label='$\\pm$ 1 std')\n",
    "        \n",
    "    plt.xticks(run_values)\n",
    "    plt.tick_params(axis='both', which='major', labelsize=12)\n",
    "    plt.ylim(0,1)\n",
    "    \n",
    "    plt.xlabel('Iterations of Genetic Algorithm', fontsize=16)\n",
    "    plt.ylabel('Abs(Spearman score)', fontsize=16)\n",
    "    \n",
    "    plt.legend(loc='center left', bbox_to_anchor=(1.05, 0.5),\n",
    "          ncol=1, fancybox=True, shadow=False, fontsize=12)\n",
    "    \n",
    "    plt.savefig(output_folder + '/GA3-Runs.svg',\n",
    "                dpi=300,\n",
    "                bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Distributions of PPIs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/earmingol/opt/anaconda3/envs/c2c_dev/lib/python3.7/site-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
      "  warnings.warn(msg, FutureWarning)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "threshold = np.percentile(ga_selected_ppi_data['score'], 90)\n",
    "\n",
    "ax = sns.distplot(ga_selected_ppi_data['score'],\n",
    "                  bins=50,\n",
    "                  kde=False,\n",
    "                  label='All Fractions',\n",
    "                  hist_kws={'alpha' : 0.2})\n",
    "\n",
    "ax = sns.distplot(ga_selected_ppi_data.loc[ga_selected_ppi_data['score'].values > 0.]['score'],\n",
    "                  bins=50,\n",
    "                  kde=False,\n",
    "                  ax=ax,\n",
    "                  label='Non-Zero Fractions',\n",
    "                  hist_kws={'alpha' : 0.2})\n",
    "\n",
    "ax.legend(loc='center left', bbox_to_anchor=(1.05, 0.5),\n",
    "          ncol=1, fancybox=True, shadow=False, fontsize=12)\n",
    "\n",
    "plt.axvline(x=threshold, c='r', ls='--')\n",
    "plt.tick_params(axis='both', which='major', labelsize=12)\n",
    "plt.xlim(0,1)\n",
    "\n",
    "plt.xlabel('Fraction of GA-Selected Times', fontsize=16)\n",
    "plt.ylabel('Frequency', fontsize=16)\n",
    "plt.savefig(output_folder + '/GA3-Distribution.svg',\n",
    "                dpi=300,\n",
    "                bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.45599999999999996"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "threshold"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conserved LR pairs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import pairwise_distances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "lrs = symbol_lrs.apply(lambda row: row['Ligand_symbol'] + '_' + row['Receptor_symbol'], axis=1)\n",
    "lrs2 = symbol_lrs.apply(lambda row: row['Ligand_symbol'] + ' / ' + row['Receptor_symbol'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "lr_dict = pd.DataFrame()\n",
    "lr_dict['symbol'] = symbol_lrs.apply(lambda row: row['Ligand_symbol'] + '_' + row['Receptor_symbol'], axis=1)\n",
    "lr_dict['WB'] = symbol_lrs.apply(lambda row: row['Ligand_WB'] + '_' + row['Receptor_WB'], axis=1)\n",
    "lr_dict.set_index('symbol', inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Presence**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame(np.array(selected_ppis), columns=lrs).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>arf-1.2_pld-1</th>\n",
       "      <th>daf-7_sma-6</th>\n",
       "      <th>srp-6_F14B4.1</th>\n",
       "      <th>adm-2_pat-3</th>\n",
       "      <th>adm-2_pat-2</th>\n",
       "      <th>let-756_sdn-1</th>\n",
       "      <th>let-756_ver-3</th>\n",
       "      <th>let-756_ver-1</th>\n",
       "      <th>srp-2_F14B4.1</th>\n",
       "      <th>srp-1_try-1</th>\n",
       "      <th>...</th>\n",
       "      <th>emb-9_ddr-2</th>\n",
       "      <th>mec-1_mec-4</th>\n",
       "      <th>mec-5_mec-4</th>\n",
       "      <th>mec-9_mec-4</th>\n",
       "      <th>mec-1_mec-10</th>\n",
       "      <th>mec-5_mec-10</th>\n",
       "      <th>mec-9_mec-10</th>\n",
       "      <th>shn-1_ced-1</th>\n",
       "      <th>col-119_daf-4</th>\n",
       "      <th>col-19_daf-4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 245 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   arf-1.2_pld-1  daf-7_sma-6  srp-6_F14B4.1  adm-2_pat-3  adm-2_pat-2  \\\n",
       "0              0            0              0            0            0   \n",
       "1              0            0              0            0            0   \n",
       "2              0            0              0            0            0   \n",
       "3              0            0              0            0            0   \n",
       "4              0            0              0            0            0   \n",
       "\n",
       "   let-756_sdn-1  let-756_ver-3  let-756_ver-1  srp-2_F14B4.1  srp-1_try-1  \\\n",
       "0              0              1              0              0            1   \n",
       "1              0              1              1              0            0   \n",
       "2              0              0              1              0            0   \n",
       "3              0              1              1              0            0   \n",
       "4              0              0              1              0            1   \n",
       "\n",
       "   ...  emb-9_ddr-2  mec-1_mec-4  mec-5_mec-4  mec-9_mec-4  mec-1_mec-10  \\\n",
       "0  ...            0            0            0            0             0   \n",
       "1  ...            0            0            1            0             1   \n",
       "2  ...            0            0            1            0             1   \n",
       "3  ...            0            0            1            0             1   \n",
       "4  ...            0            0            0            0             1   \n",
       "\n",
       "   mec-5_mec-10  mec-9_mec-10  shn-1_ced-1  col-119_daf-4  col-19_daf-4  \n",
       "0             1             1            0              1             0  \n",
       "1             1             0            0              1             0  \n",
       "2             1             1            0              1             1  \n",
       "3             1             1            0              1             1  \n",
       "4             0             1            0              1             1  \n",
       "\n",
       "[5 rows x 245 columns]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/earmingol/opt/anaconda3/envs/c2c_dev/lib/python3.7/site-packages/seaborn/matrix.py:654: UserWarning: Clustering large matrix with scipy. Installing `fastcluster` may give better performance.\n",
      "  warnings.warn(msg)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.clustermap(df,\n",
    "               metric='jaccard',\n",
    "               method='complete',\n",
    "               col_cluster=True,\n",
    "               row_cluster=True,\n",
    "               cmap='Blues',\n",
    "               figsize=(16, 9)\n",
    "              )\n",
    "plt.savefig(output_folder + '/GA3-LRs-Runs.svg',\n",
    "                dpi=300,\n",
    "                bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Co-occurrence**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "co = pd.DataFrame(np.zeros((len(lrs), len(lrs))), columns=lrs, index=lrs)\n",
    "for i, lr in enumerate(lrs):\n",
    "    v1 = df[lr]\n",
    "    for j, lr2 in enumerate(lrs[i:]):\n",
    "        v2 = df[lr2]\n",
    "        if sum(v1.values | v2.values) == 0:\n",
    "            val = 0.0\n",
    "        else:\n",
    "            val = sum(v1.values & v2.values) / sum(v1.values | v2.values)\n",
    "        \n",
    "        co.at[lr, lr2] = val\n",
    "        co.at[lr2, lr] = val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>arf-1.2_pld-1</th>\n",
       "      <th>daf-7_sma-6</th>\n",
       "      <th>srp-6_F14B4.1</th>\n",
       "      <th>adm-2_pat-3</th>\n",
       "      <th>adm-2_pat-2</th>\n",
       "      <th>let-756_sdn-1</th>\n",
       "      <th>let-756_ver-3</th>\n",
       "      <th>let-756_ver-1</th>\n",
       "      <th>srp-2_F14B4.1</th>\n",
       "      <th>srp-1_try-1</th>\n",
       "      <th>...</th>\n",
       "      <th>emb-9_ddr-2</th>\n",
       "      <th>mec-1_mec-4</th>\n",
       "      <th>mec-5_mec-4</th>\n",
       "      <th>mec-9_mec-4</th>\n",
       "      <th>mec-1_mec-10</th>\n",
       "      <th>mec-5_mec-10</th>\n",
       "      <th>mec-9_mec-10</th>\n",
       "      <th>shn-1_ced-1</th>\n",
       "      <th>col-119_daf-4</th>\n",
       "      <th>col-19_daf-4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>arf-1.2_pld-1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>daf-7_sma-6</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.011494</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.030303</td>\n",
       "      <td>0.011236</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.010753</td>\n",
       "      <td>0.034483</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>srp-6_F14B4.1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>adm-2_pat-3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>adm-2_pat-2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 245 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               arf-1.2_pld-1  daf-7_sma-6  srp-6_F14B4.1  adm-2_pat-3  \\\n",
       "arf-1.2_pld-1            0.0          0.0            0.0          0.0   \n",
       "daf-7_sma-6              0.0          1.0            0.0          0.0   \n",
       "srp-6_F14B4.1            0.0          0.0            0.0          0.0   \n",
       "adm-2_pat-3              0.0          0.0            0.0          0.0   \n",
       "adm-2_pat-2              0.0          0.0            0.0          0.0   \n",
       "\n",
       "               adm-2_pat-2  let-756_sdn-1  let-756_ver-3  let-756_ver-1  \\\n",
       "arf-1.2_pld-1          0.0            0.0            0.0       0.000000   \n",
       "daf-7_sma-6            0.0            0.0            0.0       0.011494   \n",
       "srp-6_F14B4.1          0.0            0.0            0.0       0.000000   \n",
       "adm-2_pat-3            0.0            0.0            0.0       0.000000   \n",
       "adm-2_pat-2            0.0            0.0            0.0       0.000000   \n",
       "\n",
       "               srp-2_F14B4.1  srp-1_try-1  ...  emb-9_ddr-2  mec-1_mec-4  \\\n",
       "arf-1.2_pld-1            0.0          0.0  ...          0.0          0.0   \n",
       "daf-7_sma-6              0.0          0.0  ...          0.0          0.0   \n",
       "srp-6_F14B4.1            0.0          0.0  ...          0.0          0.0   \n",
       "adm-2_pat-3              0.0          0.0  ...          0.0          0.0   \n",
       "adm-2_pat-2              0.0          0.0  ...          0.0          0.0   \n",
       "\n",
       "               mec-5_mec-4  mec-9_mec-4  mec-1_mec-10  mec-5_mec-10  \\\n",
       "arf-1.2_pld-1          0.0          0.0      0.000000      0.000000   \n",
       "daf-7_sma-6            0.0          0.0      0.030303      0.011236   \n",
       "srp-6_F14B4.1          0.0          0.0      0.000000      0.000000   \n",
       "adm-2_pat-3            0.0          0.0      0.000000      0.000000   \n",
       "adm-2_pat-2            0.0          0.0      0.000000      0.000000   \n",
       "\n",
       "               mec-9_mec-10  shn-1_ced-1  col-119_daf-4  col-19_daf-4  \n",
       "arf-1.2_pld-1           0.0          0.0       0.000000      0.000000  \n",
       "daf-7_sma-6             0.0          0.0       0.010753      0.034483  \n",
       "srp-6_F14B4.1           0.0          0.0       0.000000      0.000000  \n",
       "adm-2_pat-3             0.0          0.0       0.000000      0.000000  \n",
       "adm-2_pat-2             0.0          0.0       0.000000      0.000000  \n",
       "\n",
       "[5 rows x 245 columns]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "co.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "col_bins = 10\n",
    "data = co.loc[(co != 0).any(), (co != 0).T.any()]\n",
    "colors = ga_selected_ppi_data['score']\n",
    "colors.index = ga_selected_ppi_data.apply(lambda row: row['A'] + '_' + row['B'], axis=1)\n",
    "colors = colors.loc[lr_dict.loc[co.columns, 'WB']]\n",
    "colors.index = co.columns\n",
    "colors_df = colors.copy()\n",
    "colors_df.index = lrs2\n",
    "colors = colors.apply(lambda x: plt.get_cmap(\"YlOrRd\",col_bins)(x))\n",
    "colors.name = 'Presence in Optimal Lists'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "colors_df.to_frame().to_csv(output_folder + '/Celegans-GA-ICELLNET-Run-Ratios.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "col-119 / daf-4    0.93\n",
       "lin-44 / mig-1     0.92\n",
       "mec-5 / mec-10     0.89\n",
       "let-756 / ver-1    0.87\n",
       "mec-9 / mec-10     0.85\n",
       "                   ... \n",
       "tig-2 / daf-4      0.00\n",
       "sfrp-1 / lin-18    0.00\n",
       "tig-3 / sma-6      0.00\n",
       "tig-2 / sma-6      0.00\n",
       "unc-52 / pat-2     0.00\n",
       "Name: score, Length: 245, dtype: float64"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "colors_df.sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "colors_export = colors.to_frame()\n",
    "colors_export.columns = ['Color']\n",
    "colors_export.index = colors_df.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "colors_export.to_csv(output_folder + '/Celegans-GA-ICELLNET-Run-Ratios-Color.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(110, 110)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "c_input = data\n",
    "D = pairwise_distances(c_input)\n",
    "linkage = hc.linkage(D, method='ward', optimal_ordering=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(110, 110)"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "D.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlIAAABsCAYAAAChMFyHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8GearUAAAaYElEQVR4nO3deZwcRfnH8c/DfYMQDCiSgCKCXAqCikBQQAERRX6AILCA+JMIiIqoeAUPEPihoKIcIiuInBJEQINGw30JqFxCMAlCOOQm5E54fn88NUmn07M7Uzuzs7t+36/XvHa3p7q7qo/qp6ura83dEREREZHmLdHpDIiIiIgMVgqkRERERDIpkBIRERHJpEBKREREJJMCKREREZFMCqREREREMimQkj4xswlmNqXT+WiGme1oZreb2TQzczPr6nSe/luZ2ci0D8Z0Oi9DhZl1m1nHx7UxsylmNqFD627bNtAxK2UKpAYQMxuVTtB6n3d3KF/HDJVgw8xeB1wJrAh8ETgQuLGjmZK2MrPtzOwiM5tsZjPNbIaZPWpmvzazj5qZ9TDvHencO68/89wbM+sys2M6nY+hKgVLY8xsi07nRQY+04CcA4eZjQL+AlwMXFeR5A/u/ly/Zoq4swSmuPuoiu+WIY6j2f2drxxmtgswDvi4u1/Z6fz8t0tBzLLAPHef1+JlLwH8BDgCeAK4DHgEeA1YD/gQ8A7geHc/qWL+TYD7gH8Bw4G13H16K/OYK7X0jHT3kRXfLQ0s6e6z+jtfpXxMoU690Q/r7gYOdve6QXIv848i6uJD3L279F3bjlkZnJbqdAak0j3u/qtGE5vZksCy7j6jjXmq5O5z+nudfbRW+vlCf63QzFZ292n9tb7BxONOrl0X/G8SQdRFwGEVwf7xZrYj8IY68x8GTAM+CdwG7AOc36a8toy7zwXmdjofQ1Wbj1kZjNxdnwHyAUYBDhzbQ5qulGYn4BvE3fJcoCt9vwtwKTAJmAm8BFwP7FBneW8hLg5PAHOAJ4HfAlum773OZ2T6fgJx11le7vbAH4GXUz7uIS5m5XQTgCnExexi4EVgBtFq9NYmtt1mwFjgeaKSexA4jrgzr6WZUlWWXpbbndKtCVyQlj8dGA+8s5R2ZEo7BtgXuDuVvbuQZqe0P15K+fwH8JmK9b4X+D3wdEo3lWilfHcp3arAycCjwGzg2bQd169z3LwfODYdN7OJFpqD65R9R+DawjadBJwHDCul2xe4mQg6ZgB3AHs3uN8WbLM62/HDwF1p/U8BpwJLNbDc16dtP4m4yWj2XFwmbctfpL/vAW7OWM6KwEmF7f10Oo5G1Dn3u4Cj0n6ZlX4eVUpbeRwDo4rHbJ3jeI30+3Npf11FtLQBfBp4KK33n8CeFeUZnY7fqUR98RTwK1J9UJHPCQ1up4OAO4nzYnrabxcBa2bWK1XbYALVddUixyALz5XyZ0K9YzZNXwr4MlH3zCLOm7HApj3UE70e38DbgcvTNq8dQ38Bdm/2eNSnPR+1SA1MK5jZsNK02b5oq8b/AUsD5wKvAA+n6V3A6kRl/QTwRuBTwHgz29Hdb6otwMy2IgKCpYkL5P1p3h2IC/ndRB+iHxIV7/cK63+2XubNbA+iAnkaOI2osPcDfm5m67v710qzrEj0U7odOJ547PI54Ldmtom7z6+3rkI5biACyjPTevcgAozNgQNS0mOAXYkLxonERaNRfyBascYQrVpHAjeY2Xvc/f5S2o8CRwM/A84i9g9m9un09+3EtpwO7Az8zMze7O5fSuk2JC4WTwNnAM8Qj5bel8pze0q3KnArsC7wC+ABYG3iYneHmW3l7o+V8nYisDxwNlEpHwF0m9mj7n5LLZGZ/W/K/9T087G0nj2AdYjjATP7LvC1tH2+QTw2+xhwuZkd6e5nNrqBK+yWynJWKt+eRBD4YipHT3YHlgMu9LzHznsCw4Bfpr+7gTPMbEN3f7juXAXpEds4YFvgCuJc2IDY5ruk/fNEabajiOPrbOK8+QTwIzNb3d1PSGmOIYKzYcDnC/M2cjz/gagXvkncRB0NjDWzK4nz4jzion40cIWZvdXdJxfmP5Y4/n5EnA+bEPXL+81sU3d/voE8LMLMDiS2800pXzOBNxH7//WkuiajXsl1I3F8HQ+ck/IFcR725CKi1fKPxDmzFvBZ4DYz287d7y2l7/X4NrM1gD+n9GcR5+EwYCtgG+JGRzqt05GcPgs/LLwrrfpcktJ0pb8fBlaoWMaKFdOGExe+6wrTjAicZgGbVcyzROH3KdS5s6R0lwcsSZzsLwFvKExfBrgFmA9sUJrfgeNKy/1Smv7BBrbbLcC8YjlS+S5Ly/hAYXpt+41qcJ90p/RXkvoUpulbEkHDHwrTRqa0c4GNSstZO23rX1es44y0XdZPfx+dlrN1L3k7g7jobF6aPoII3roryn0vsExh+huJgOriwrR10rQHgdXqHRvAO9MyT6xIc1XKw8q9lKG2zcZUTJtOoaWjcMw+1cB+Oy0t42MV361KXIxqn6oy/h6YXNvnKd0c4OQmzufDUx5OKU3fPU2/sDBtVJo2DVindN7cmY6p4vQJVLSuFI/ZOsfxmaXpP0jT/w2sUpi+WZp+Uil9Vf3yAarP4Sk00CJFnFuv0ENLI83XK1XboHKb1TkGa/ujq8H0O6dpl7JoPbE5UTfdlHN8Ax9Jafdp9LjTp/8/emtvYDqHODGLn++W0vzMK/pEeaEzrJmtlO5o5hOPW7YpJN2CaDI+393/UbGc1zLzviWphcTdnywsbw5wCvGm6J6leV4j7nCLandhG/S0MjN7PdF6dnWxHB61UK0F7WNNlqHKKWmZteXfTdx57mRmK5XSXuvu5daBvYkOqueZ2bDiB/gdsV12SmlfTj/3NLPlqjKTOrweQNw9Ty0tbzrRarBLxaw/9UK/NnefSjw+Km7n/yEuUCe4+0vlBRSOjQOISv6XFWW6GlgZeE9V/ht0lbtPKazXiUcaa1Vs87JV0s9XKr4bT7Ry1D43F780szcR2+6C2j73eMnjWuAgM2u0Jf9jxLG9SEd2d78W+Buxf8t18EVeaKVK++qHxGOjPRpcb09OL/1da225wN0XbKt0Lr1C6fyr1S9mtoSZrZr29d+JY7ZYvzTjZWAFYPce3qDMqVf6U62O+V6pnvg7cX6/z8zWLM3TyPFdqwt2NbNVkAFJj/YGponu/qde0jxSNdHM3kwEEB8EVit97YXfaxVkubm5r9ZLPx+o+K42bf3S9Cd98TeMao8I1ujD+h4iLmTl9eWoemzyIHHBHVFaf9W+2Sj97Gm/Dk8/LyE6OB8PfN7MbiceEV3iCx/VrUlsm12o/5i1KhieVDHteaIMNY0eGxsRd9H/7CHN8B6+6029vEKU/dUe5q0FBVUXn9GF6VUvdXQRF+ZbzOwthel/Jh7b7kYEiqSL45KFNPPdvbY/1iOO7Rcr1vEAcTMzDPhPYXq94wxacxyXt2ktb5PLCdN3i5x/ZvZ+4vHbNsSj06LXZebpRKLv01XA82Z2A9EieKkv7M6QU6/0p/WI861q/z1AHDfrsei52uvx7e43mNkFxDF5gJndRdQhl7r7gxXzSwcokBq8FmuNSncxNxJ9jk4nXt2eRpzgXyU6Gg9EPfWBynp9ucOq3p6sleMgolNplUkAHn16djazrYmAeHvg28AYM9vf3ccWlvcnoi9Yo+pt65ztbERwvmsPy6268DWqL8dFrd/aFkS/mgXc/c4FCzFbJIBPLSKHpD/H1Vn2oaRAiugoXAxCHyMe3QxIXr+/Ya/HhZm9i+ho/ijwFSL4mknqekDmuITuPtHMNiYeEX6A6KN5LnCCmW3v7v/KWW7VqupM79R1sKHj290PNrNTifNsO2L8u6+Z2THu/pM251EaoEBqaPkA8fbboe5+fvGL1Cm4qNZq0siAc/UqoCq1u6y3V3y3cSlNK9TupKvW9zaicm/F+jYidfIu2JioDMsduqtMTD+fa6C1EVhwwb8TFjxuupd4xDuWuLN9iejX0tDymlA8NipbPpOJxFhM/654lNlp1xJ90g40sxO98Q7nOxItB6cTfW/KPgF8xMyGu/szxOPN5Qvfzyz8Pgn4kJmtVvGIdGOi1aw8LtxGLK7qvGnmnGyV/YnWt1290AHdzFYkvzUKWHDzcF36YGa7EfvwC0SH7VbUKy8QjwjLqlqymt2+k4i6ZiPiTdyiWv6qWv0a4vFCy/3AqWa2GtFV4/tmdmbxUaJ0hvpIDS21O5xF7tbTIJTl/gt/J1oLDjWzxSqnUl+FV4m3+RpxD9Fx9RAzq43ZVHuDqdaB/LcNLqtX7v4f4s21PdIAirX1GdEKB6UWiUzHFbeJmb2T6NM03t17esRUcxnRgfsEM1u+/GXqb7Js+r38xibEm1bPkvZD6qd0EbC1me1dtcLUfyzHFUTH6m9V9csobIcL088T01hm5XR9eazXJ+m4OIUIin5R27YVyi1bhxHn0YnufkX5Q/TlW4poWcTdb3H3PxU+xeDrKqKO/coiKzTblRgI9OqKvogHmNk6hbTLEG/mzQeuKaR7FXhdD32K2qGyfiEeQWdfS+oc7/ekn6sX/u5rvfIIsHJq6a3NvwSLvvlYUzunG633rko/v1qqJzYhOozfXHjk2zAzW73cjy4F5ZOJfmWVfSilf6lFami5mfRqsJmNJC6+WxBDGNwHbFpL6O5uZocQHW/vtPgXGPcT/ap2IF6T/nFKfjtwmJl9h4X9jn7nFaM8u/t8MzuSCF7uMrNziMeL+wLvJi5QE8vz9dHniOEPbjKz2vAHHyYei/3a3ce3YB0jgHFmdjXxBt6RROvDlxqZ2d2fMLMjgJ8DD5nZhURL1prEfvkocec6Bfh6Cn6vIb05RnQ0fhsRHNR8jXi1/jIzu4zYT3NSXncjhq/oaragKa/HEENJ3Jf6aDxGvOG3J/Fo62/ufpfF/xsbA/zNzC4nxiFbm7jz343otN4pJxCvz38G2D5to9rQBesQF7h1SQFKutPfi3jDqt5F7yaiT9OhxJg/PekGDga+nM7HG4khB0YTr9IfXzHPI8TQFWcR583+wLuA77j744V0txPH+E/M7FYiyPlzCiDbZSwRdFyXzus5xIswm7F4y1ozrjezl4ht+zhRB3WR3myEltUr5xCPxcaa2Rkp/3tTfR18MC1/tJnNIFp//+Puf65Ii7v/MR1f+xEB7jUsHP6gNpxEjoOIfpJjiUeqc4n6+YPAZe4+s6eZpZ/05yuC+vT8obkBOUfV+X4zIgh6kagIJhDP1bupGHwS2JDocPs0CwfkvIrCYJPExeg3RNP4azQ2IOcOxFttrxAVyb30MCBnxfSRVAx618N22Tzl+wWi5echSgNyNrL9KpbbzcIBOS8kOoPOIDoeb9lsnonAZyxxMa5t778QFfxyhePgUiKompnKdAcxXo+VlrcCMX7TfSnttFT2c4FtGil3D/tgFxYOflgbkPNcYI1Sut2J/kS1bf840Vl4sYFGG9nPPW1HImhbcPw1uA93IFrvpqRyzCQGyLyYCAxrQxx8Ni37qF6Wd3ZK994G1l0bkHNS2t//ScfRiDrnfhdx0Z2YtuVE4HMVy12BGPPpGSKIWrBv6WH4g4rlLFhvxXdTKA1fQAT8dxNvhj5H9I1at07axabV2UaHs3DctNogn9cBO9bZl43UK/XKuxvxxuRs4tw7magDFzveUtp70nqcxgfkfCgt/wWiTqo7IGdvxzdxI/xLIoiansr9d6K+aHqgWX3a89H/2hPpgfXxf3aJNMJ6+N9uIjKwqY+UiIiISCYFUiIiIiKZFEiJiIiIZFIfKREREZFMapESERERydSycaSGrbGKjxxR/p+MQ8j8Ob2nGczmDeHyzWl0UOtBam7u/5ceJGb19J80Bj+fObT337xZQ7d8c+d1OgftNfu1of2y8iRmP+fufQ5cWhZIjRyxJn+94ZTeEw5S/koj/wVkEHvh353OQfs8PqXTOWgrn/pK74kGMX+0kYHjB6/5903rPdEg9sLDVf96cmh48pnFBvQfUiZPX7rTWWirvXmkJRd2PdoTERERyaRASkRERCSTAikRERGRTAqkRERERDIpkBIRERHJpEBKREREJJMCKREREZFMCqREREREMimQEhEREcmkQEpEREQkkwIpERERkUwKpEREREQyKZASERERyaRASkRERCSTAikRERGRTAqkRERERDIpkBIRERHJpEBKREREJJMCKREREZFMCqREREREMimQEhEREcmkQEpEREQkkwIpERERkUwKpEREREQyKZASERERyaRASkRERCSTAikRERGRTAqkRERERDIpkBIRERHJpEBKREREJJMCKREREZFMCqREREREMimQEhEREcmkQEpEREQkkwIpERERkUwKpEREREQyKZASERERyaRASkRERCSTAikRERGRTAqkRERERDIpkBIRERHJpEBKREREJJMCKREREZFMCqREREREMimQEhEREcmkQEpEREQkkwIpERERkUwKpEREREQyKZASERERyaRASkRERCSTAikRERGRTAqkRERERDKZu7dmQWbPAo+1ZGEiIiIi7TXC3dfs60JaFkiJiIiI/LfRoz0RERGRTAqkRERERDIpkBIRERHJpEBKREREJJMCKREREZFMDQdSZjbazCab2Swzu9vMtusl/Q4p3Swzm2Rmn+l7dtunmfKZ2dpm9msz+6eZzTez7n7MapYmy7eXmV1vZs+a2TQzu8PMPtKf+W1Gk2XbwcxuNbPnzWxm2ofH9md+m9XsuVeY731mNs/M7m93Hvuiyf03ysy84vO2/sxzMzLqzmXM7Ntpntlm9m8zO7q/8tusJvdfd539N70/89yojH23v5n9zcxmmNnTZvYrM1urv/LbrIzyfdbMHkp158NmdlB/5bVZZra9mV1tZlPTMdbVwDybmtkNqXxTzeybZma9rszde/0A+wJzgcOBjYAfA68C69ZJvx4wPaXbKM03F/h4I+vr709G+UYCPwK6gFuB7k6XocXlOwP4CrA18BbgW8B8YLtOl6UFZdsS2A94ezpOP5mO1dGdLksryleY73XAJGAccH+ny9HC/TcKcGBjYK3CZ8lOl6VV+w+4ErgT2DnVNdsAozpdlhbtv1VL+20t4F/A+Z0uSwvKtm2qJz+f6pZ3A/cA4ztdlhaV74j0/SeA9VM9Og3Yo9NlqZPf3YATgb2BGUBXL+lXAZ4GLgM2SfNNA77Y67oazNAdwLmlaROBk+qkPxmYWJr2c+C2Tm/cVpSvlO4aBn4glV2+Qvo7gdM6XZY2le1K4OJOl6WV5Utl+hYwhoEdSDVbt4wiAqlhnc57m8q3C/DyUC1fxfzbpv353k6XpQX77ljgsdK0Q4BXO12WFpXvVuCHpWmnATd3uiwNlPXVBgKpI4BXgOUL074OTCWNuVnv0+ujPTNbhriLv7701fXAe+vM9p6K9OOArcxs6d7W2Z8yyzdotLB8KwMvtipfrdCKspnZO1LaG1qbu77LLZ+ZjQaGA99tX+76ro/7769m9pSZjTezHduSwT7KLN9HgbuAL5jZE2Y20cx+ZGYrtTGrWVpUtxwOPODut7Yyb32VWbZbgLXNbA8Lw4hWm+val9M8meVbFphVmjYT2HqgXdczvQe4yd1nFqaNA95AtAzX1UgfqWHAksAzpenPEM2yVdaqk36ptLyBJKd8g0mfy2dmnwXWAS5sbdb6LLts6SI1G/gr8FN3P6s9WeyTpstnZpsSLVGfdPf57c1en+Xsv6eIO8ePA3sBDwPjG+031s9yyrc+8D5gc6KMRwIfArrbk8U+6VPdYmarAvsA57Y+a33WdNnc/TYicLoImAM8CxhwcPuymS1n340DDjWzd6VAcSvgU8DSDLzreo56cUvtu7qWakt2ZMgws48DpwL7uvtQ+l+K2wErEf0YTjazye4+0ALFppjZssClwLHuPrnT+WkHd3+YCJ5qbjOzkcCXgJs6kacWW4J41LW/u78MYGZHAuPMbLi7lyv6weyTRHkH9XlXY2YbE/2MvkMEHWsTdefZwIDtlN2E7xABxa1EgPgM8EvgOOC1Duar4xppkXqO6EA3vDR9ONExq8rTddLPS8sbSHLKN5hkl8/M9iYquYPc/XftyV6fZJfN3Se7+33ufi7wA6Iv0UDTbPnWJjqNnp/e1psHfBN4e/p7l7bmtnmtOvfuADZoVaZaKKd8TwFTa0FU8lD6uW5rs9dnfd1/hwO/cfcXWp2xFsgp21eBO939VHf/h7uPA0YDB5rZOu3Lapamy+fuM939UGAF4lHXusAUokP2s+3KaD+qF7fUvqur10DK3ecAdxNvkBTtTESmVW6rk/6v7j63t3X2p8zyDRq55TOzfYggqsvdr2hfDvO1cN8tQTz/H1AyyjcV2BTYovA5C3g0/T6gjucW7r8tiABkQMks3y3AG0p9ot6afg6oFuG+7D8z25p4fDkQH+vllm0FIjgpqv09oMZs7Mu+c/e57v5E6jqwH3CNuw+FFqnbgO3MbLnCtJ2BJ4mAsb4Ge7zvSzzz/RRxx3sG0Qt+RPr+AuCCQvra8Aenp/SfSvMP5OEPGi5fmla7UN0IXJ1+37jTZWnR/tuPeC32cyz6mvLqnS5LC8p2FPBhogVjA+Aw4k2N73e6LK06Nkvzj2Fgv7XX7P47huiQvQExhMVJxKOwvTpdlhaVbyXgceDyVL5tgfuByztdllYen8Rb3I90Ov8t3nddqd48gujrti3x4sDdnS5Li8r3VuDAdO5tDVwCPA+M7HRZ6pRvJRZep2cQrfNbkIZ3SHXH+EL6VYmWp0uI4Q/2SteG1gx/kFYymojKZhOR7PaF7yYAE0rpdyDG0JgNTAY+0+kN2+LyecVnSqfL0Yrypb+ryjehv/PdhrIdAzxABPovp2N0NLBEp8vRqmOzNO8YBnAglbH/jiNe0Z4JvED0i9qt02Vo5f4DNiTenppBtDKeCazc6XK0sHwrExfs4zqd9zaU7ahUv8wgWkkvAtbpdDlaUT4i2Lo3le1l4Cpgw06XoYeyjaL6Otadvu+mdM0mWvRvJN5OfIp4cafHoQ/cPRKIiIiISPMG1HNbERERkcFEgZSIiIhIJgVSIiIiIpkUSImIiIhkUiAlIiIikkmBlIiIiEgmBVIiIiIimRRIiYiIiGRSICUiIiKS6f8B+gGI6lOk9poAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 720x72 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.palplot(sns.color_palette(\"YlOrRd\", col_bins))\n",
    "for i, text in enumerate(sns.color_palette(\"YlOrRd\", col_bins + 1)):\n",
    "    plt.text(i-0.5, 0.6, i/col_bins, rotation=0, horizontalalignment='center', verticalalignment='top',fontsize=14)\n",
    "plt.title('Fraction of presence in GA-optimal solutions', fontsize=18)\n",
    "plt.tick_params(axis='both', which='major', bottom=False, left=False,labelsize=16)\n",
    "plt.savefig(output_folder + '/GA-ICELLNET-Fraction-Colors.svg',\n",
    "                dpi=300,\n",
    "                bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "co_clusters = sns.clustermap(c_input,\n",
    "                             method='ward',\n",
    "                             cmap='Blues',\n",
    "                             col_linkage=linkage,\n",
    "                             row_linkage=linkage,\n",
    "                             figsize=(15, 15),\n",
    "                             col_colors=colors\n",
    "                             )\n",
    "# Title\n",
    "co_clusters.ax_col_dendrogram.set_title('Co-Ocurrence of LRs in Optimal Lists', fontsize=16)\n",
    "\n",
    "# Color bar label\n",
    "cbar = co_clusters.ax_heatmap.collections[0].colorbar\n",
    "cbar.ax.set_ylabel('Co-Occurrence', fontsize=14)\n",
    "cbar.ax.yaxis.set_label_position(\"left\")\n",
    "\n",
    "plt.savefig(output_folder + '/GA-ICELLNET-Co-Ocurrence.svg',\n",
    "                dpi=300,\n",
    "                bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "clusters = c2c.clustering.get_clusters_from_linkage(co_clusters.dendrogram_col.linkage, 2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Export Selected LRs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Selected by Co-occurrence**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "records = []\n",
    "selected_cluster = np.argmin([len(c) for c in clusters.values()]) + 1\n",
    "\n",
    "for idx, row in lr_dict.loc[co_clusters.data.index[clusters[selected_cluster]]].iterrows():\n",
    "    records.append((*row['WB'].split('_'), *idx.split('_')))\n",
    "    \n",
    "ga_export = pd.DataFrame.from_records(records, \n",
    "                                      columns=['Ligand_WB', 'Receptor_WB', 'Ligand_symbol', 'Receptor_symbol'])\n",
    "\n",
    "ga_export = ga_export.assign(score=1.0)\n",
    "ga_export = ga_export.merge(symbol_lrs, how='inner')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Ligand_WB</th>\n",
       "      <th>Receptor_WB</th>\n",
       "      <th>Ligand_symbol</th>\n",
       "      <th>Receptor_symbol</th>\n",
       "      <th>score</th>\n",
       "      <th>index</th>\n",
       "      <th>LR Function</th>\n",
       "      <th>L Function</th>\n",
       "      <th>R Function</th>\n",
       "      <th>Ligand_desc_WormBase_versionWS277</th>\n",
       "      <th>Receptor_desc_WormBase_versionWS277</th>\n",
       "      <th>Pair_confidence</th>\n",
       "      <th>Location_type</th>\n",
       "      <th>Confidence_evidence</th>\n",
       "      <th>Location_evidence</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>WBGene00002881</td>\n",
       "      <td>WBGene00006896</td>\n",
       "      <td>let-756</td>\n",
       "      <td>ver-3</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6</td>\n",
       "      <td>Canonical RTK-Ras-ERK signaling</td>\n",
       "      <td>FGF</td>\n",
       "      <td>Tyrosine-protein kinase (GF-Receptor)</td>\n",
       "      <td>Is an ortholog of human FGF16 (fibroblast grow...</td>\n",
       "      <td>Is an ortholog of human FLT1 (fms related rece...</td>\n",
       "      <td>Putative based on literature</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase: let-756 is found in the cytoplasm;ex...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>WBGene00002881</td>\n",
       "      <td>WBGene00006894</td>\n",
       "      <td>let-756</td>\n",
       "      <td>ver-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>7</td>\n",
       "      <td>Canonical RTK-Ras-ERK signaling</td>\n",
       "      <td>FGF</td>\n",
       "      <td>fms-related tyrosine kinase 1</td>\n",
       "      <td>Is an ortholog of human FGF16 (fibroblast grow...</td>\n",
       "      <td>Is an ortholog of human FLT1 (fms related rece...</td>\n",
       "      <td>Putative based on literature</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase: let-756 is found in the cytoplasm;ex...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>WBGene00003029</td>\n",
       "      <td>WBGene00003238</td>\n",
       "      <td>lin-44</td>\n",
       "      <td>mig-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>25</td>\n",
       "      <td>Wnt signaling</td>\n",
       "      <td>Wnt</td>\n",
       "      <td>Frizzled</td>\n",
       "      <td>Is an ortholog of human WNT10A (Wnt family mem...</td>\n",
       "      <td>Is predicted to have Wnt-activated receptor ac...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...</td>\n",
       "      <td>wormbase: lin-44 is said to be located in the ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>WBGene00002280</td>\n",
       "      <td>WBGene00003930</td>\n",
       "      <td>let-2</td>\n",
       "      <td>pat-3</td>\n",
       "      <td>1.0</td>\n",
       "      <td>26</td>\n",
       "      <td>Cell adhesion</td>\n",
       "      <td>Collagen</td>\n",
       "      <td>Integrin</td>\n",
       "      <td>Is an ortholog of human COL4A5 (collagen type ...</td>\n",
       "      <td>Is an ortholog of human ITGB1 (integrin subuni...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>ECM-component</td>\n",
       "      <td>10.1083/jcb.201903124</td>\n",
       "      <td>Wormbase: let-2 is located in the extracellula...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>WBGene00001164</td>\n",
       "      <td>WBGene00006868</td>\n",
       "      <td>efn-3</td>\n",
       "      <td>vab-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>27</td>\n",
       "      <td>Eph receptor signaling</td>\n",
       "      <td>Ephrin</td>\n",
       "      <td>EphR</td>\n",
       "      <td>Is an ortholog of human EFNB1 (ephrin B1); EFN...</td>\n",
       "      <td>Is an ortholog of human EPHA1 (EPH receptor A1...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>WormBook</td>\n",
       "      <td>wormbase: efn-3 is located in the plasma membr...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>WBGene00000674</td>\n",
       "      <td>WBGene00016104</td>\n",
       "      <td>col-99</td>\n",
       "      <td>ddr-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>35</td>\n",
       "      <td>Cell adhesion</td>\n",
       "      <td>Collagen</td>\n",
       "      <td>Discoidin domain receptor</td>\n",
       "      <td>Is an ortholog of human COL13A1 (collagen type...</td>\n",
       "      <td>Is an ortholog of human DDR1 (discoidin domain...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>ECM-component</td>\n",
       "      <td>DOI: 10.1016/j.mcn.2018.03.003</td>\n",
       "      <td>wormbase: col-99 is located in the extracellul...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>WBGene00003395</td>\n",
       "      <td>WBGene00003397</td>\n",
       "      <td>mom-2</td>\n",
       "      <td>mom-5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>38</td>\n",
       "      <td>Wnt signaling</td>\n",
       "      <td>Wnt</td>\n",
       "      <td>Frizzled</td>\n",
       "      <td>Exhibits receptor tyrosine kinase binding acti...</td>\n",
       "      <td>Is an ortholog of human FZD1 (frizzled class r...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...</td>\n",
       "      <td>https://www.ncbi.nlm.nih.gov/gene?Db=gene&amp;Cmd=...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>WBGene00006746</td>\n",
       "      <td>WBGene00007764</td>\n",
       "      <td>unc-6</td>\n",
       "      <td>C27B7.7</td>\n",
       "      <td>1.0</td>\n",
       "      <td>44</td>\n",
       "      <td>Cell migration</td>\n",
       "      <td>Netrin (dorsal-ventral)</td>\n",
       "      <td>Ig domain - Receptor</td>\n",
       "      <td>Is an ortholog of human NTN1 (netrin 1). Is pr...</td>\n",
       "      <td>Is an ortholog of human DSCAM (DS cell adhesio...</td>\n",
       "      <td>Putative based on literature</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase: unc-6 is found to be located in axon...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>WBGene00001700</td>\n",
       "      <td>WBGene00004208</td>\n",
       "      <td>grd-11</td>\n",
       "      <td>ptc-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>50</td>\n",
       "      <td>Hedgehog signaling</td>\n",
       "      <td>DHH</td>\n",
       "      <td>PTCH1</td>\n",
       "      <td>Is an ortholog of human DHH (desert hedgehog s...</td>\n",
       "      <td>Is an ortholog of human PTCH1 (patched 1) and ...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>DOIs: 10.1186/1471-2164-7-280; 10.1016/j.ydbio...</td>\n",
       "      <td>wormbase: ptc-1 is a receptor that is membrane...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>WBGene00000857</td>\n",
       "      <td>WBGene00003238</td>\n",
       "      <td>cwn-1</td>\n",
       "      <td>mig-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>60</td>\n",
       "      <td>Wnt signaling</td>\n",
       "      <td>Wnt</td>\n",
       "      <td>Frizzled</td>\n",
       "      <td>Is an ortholog of human WNT4 (Wnt family membe...</td>\n",
       "      <td>Is predicted to have Wnt-activated receptor ac...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...</td>\n",
       "      <td>wormbase: annotated as found in the extracellu...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>WBGene00006750</td>\n",
       "      <td>WBGene00006765</td>\n",
       "      <td>unc-10</td>\n",
       "      <td>unc-29</td>\n",
       "      <td>1.0</td>\n",
       "      <td>62</td>\n",
       "      <td>Neuronal communication</td>\n",
       "      <td>Organization of presynaptic sites</td>\n",
       "      <td>Cholinergic Receptor</td>\n",
       "      <td>Is an ortholog of human RIMS1 (regulating syna...</td>\n",
       "      <td>Is an ortholog of several human genes includin...</td>\n",
       "      <td>Putative based on literature</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase: annotated as membrane-bound, not fou...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>WBGene00006926</td>\n",
       "      <td>WBGene00004374</td>\n",
       "      <td>vit-2</td>\n",
       "      <td>rme-2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>63</td>\n",
       "      <td>Receptor-Mediated-Endocytosis</td>\n",
       "      <td>rme-2 ligand</td>\n",
       "      <td>LDL receptor</td>\n",
       "      <td>Is predicted to have lipid transporter activit...</td>\n",
       "      <td>Exhibits low-density lipoprotein particle rece...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>WormBook</td>\n",
       "      <td>wormbase: located in cytoplasmic vesicle, veis...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>WBGene00002248</td>\n",
       "      <td>WBGene00003930</td>\n",
       "      <td>lam-3</td>\n",
       "      <td>pat-3</td>\n",
       "      <td>1.0</td>\n",
       "      <td>64</td>\n",
       "      <td>Cell adhesion</td>\n",
       "      <td>Laminin</td>\n",
       "      <td>Integrin</td>\n",
       "      <td>Is an ortholog of human LAMA2 (laminin subunit...</td>\n",
       "      <td>Is an ortholog of human ITGB1 (integrin subuni...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>ECM-component</td>\n",
       "      <td>WormBook</td>\n",
       "      <td>wormbase: located in basement membrane; https:...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>WBGene00001188</td>\n",
       "      <td>WBGene00003238</td>\n",
       "      <td>egl-20</td>\n",
       "      <td>mig-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>76</td>\n",
       "      <td>Wnt signaling</td>\n",
       "      <td>Wnt</td>\n",
       "      <td>Frizzled</td>\n",
       "      <td>Is an ortholog of human WNT16 (Wnt family memb...</td>\n",
       "      <td>Is predicted to have Wnt-activated receptor ac...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...</td>\n",
       "      <td>wormbase: located in extracellular space; enab...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>WBGene00001103</td>\n",
       "      <td>WBGene00003001</td>\n",
       "      <td>dsl-1</td>\n",
       "      <td>lin-12</td>\n",
       "      <td>1.0</td>\n",
       "      <td>81</td>\n",
       "      <td>Notch signaling</td>\n",
       "      <td>Notch</td>\n",
       "      <td>Notch</td>\n",
       "      <td>Exhibits Notch binding activity. Is expressed ...</td>\n",
       "      <td>Is an ortholog of human NOTCH4 (notch receptor...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>WormBook</td>\n",
       "      <td>wormbase: predicted to be integral component o...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>WBGene00003111</td>\n",
       "      <td>WBGene00004048</td>\n",
       "      <td>mab-20</td>\n",
       "      <td>plx-2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>86</td>\n",
       "      <td>Cell migration</td>\n",
       "      <td>Semaphorin (axon guidance)</td>\n",
       "      <td>Plexin (semaphorin receptor)</td>\n",
       "      <td>Is an ortholog of human SEMA4D (semaphorin 4D)...</td>\n",
       "      <td>Is an ortholog of human PLXNC1 (plexin C1) and...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>DOI: 10.1371/journal.pgen.1005950</td>\n",
       "      <td>wormbase: located in cell surface; enables sem...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>WBGene00003029</td>\n",
       "      <td>WBGene00000289</td>\n",
       "      <td>lin-44</td>\n",
       "      <td>cam-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>95</td>\n",
       "      <td>Wnt signaling</td>\n",
       "      <td>Wnt</td>\n",
       "      <td>Wnt-Receptor (Antagonistic activity)</td>\n",
       "      <td>Is an ortholog of human WNT10A (Wnt family mem...</td>\n",
       "      <td>Is an ortholog of human ROR1 (receptor tyrosin...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>DOI: 10.1242/dev.005363</td>\n",
       "      <td>wormbase: predicted to be located in extracell...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>WBGene00021594</td>\n",
       "      <td>WBGene00000900</td>\n",
       "      <td>tig-3</td>\n",
       "      <td>daf-4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>97</td>\n",
       "      <td>TGF-B signaling</td>\n",
       "      <td>BMP</td>\n",
       "      <td>TGF-B-R</td>\n",
       "      <td>Is predicted to have cytokine activity. Human ...</td>\n",
       "      <td>Is an ortholog of human ACVR2B (activin A rece...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>DOI: 10.1016/j.isci.2022.103791</td>\n",
       "      <td>wormbase: predicted to be located in extracell...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>WBGene00003395</td>\n",
       "      <td>WBGene00003007</td>\n",
       "      <td>mom-2</td>\n",
       "      <td>lin-18</td>\n",
       "      <td>1.0</td>\n",
       "      <td>101</td>\n",
       "      <td>Wnt signaling</td>\n",
       "      <td>Wnt</td>\n",
       "      <td>Ryk/Derailed</td>\n",
       "      <td>Exhibits receptor tyrosine kinase binding acti...</td>\n",
       "      <td>Is an ortholog of human RYK (receptor like tyr...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...</td>\n",
       "      <td>wormbase: predicted to be located in extracell...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>WBGene00000858</td>\n",
       "      <td>WBGene00000478</td>\n",
       "      <td>cwn-2</td>\n",
       "      <td>cfz-2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>111</td>\n",
       "      <td>Wnt signaling</td>\n",
       "      <td>Wnt</td>\n",
       "      <td>Frizzled</td>\n",
       "      <td>Is an ortholog of human WNT5A (Wnt family memb...</td>\n",
       "      <td>Is an ortholog of human FZD5 (frizzled class r...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...</td>\n",
       "      <td>wormbase: predicted to be located in extracell...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>WBGene00003029</td>\n",
       "      <td>WBGene00000478</td>\n",
       "      <td>lin-44</td>\n",
       "      <td>cfz-2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>112</td>\n",
       "      <td>Wnt signaling</td>\n",
       "      <td>Wnt</td>\n",
       "      <td>Frizzled</td>\n",
       "      <td>Is an ortholog of human WNT10A (Wnt family mem...</td>\n",
       "      <td>Is an ortholog of human FZD5 (frizzled class r...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...</td>\n",
       "      <td>wormbase: predicted to be located in extracell...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>WBGene00002280</td>\n",
       "      <td>WBGene00001184</td>\n",
       "      <td>let-2</td>\n",
       "      <td>egl-15</td>\n",
       "      <td>1.0</td>\n",
       "      <td>131</td>\n",
       "      <td>Cell adhesion</td>\n",
       "      <td>Collagen</td>\n",
       "      <td>FGF-R</td>\n",
       "      <td>Is an ortholog of human COL4A5 (collagen type ...</td>\n",
       "      <td>Is an ortholog of human FGFR1 (fibroblast grow...</td>\n",
       "      <td>Putative based on literature</td>\n",
       "      <td>ECM-component</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase; \"An extracellular matrix structural ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>WBGene00004889</td>\n",
       "      <td>WBGene00004048</td>\n",
       "      <td>smp-1</td>\n",
       "      <td>plx-2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>134</td>\n",
       "      <td>Cell migration</td>\n",
       "      <td>Semaphorin</td>\n",
       "      <td>Plexin (semaphorin receptor)</td>\n",
       "      <td>Is an ortholog of human SEMA6A (semaphorin 6A)...</td>\n",
       "      <td>Is an ortholog of human PLXNC1 (plexin C1) and...</td>\n",
       "      <td>Putative based on literature</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase, \"Predicted to be located in extracel...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>WBGene00004890</td>\n",
       "      <td>WBGene00004048</td>\n",
       "      <td>smp-2</td>\n",
       "      <td>plx-2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>135</td>\n",
       "      <td>Cell migration</td>\n",
       "      <td>Semaphorin</td>\n",
       "      <td>Plexin (semaphorin receptor)</td>\n",
       "      <td>Is an ortholog of several human genes includin...</td>\n",
       "      <td>Is an ortholog of human PLXNC1 (plexin C1) and...</td>\n",
       "      <td>Putative based on literature</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase, \"exracellular region,\" https://pubme...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>WBGene00004854</td>\n",
       "      <td>WBGene00004048</td>\n",
       "      <td>slt-1</td>\n",
       "      <td>plx-2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>136</td>\n",
       "      <td>Cell migration</td>\n",
       "      <td>Slit (dorsal nerve cord)</td>\n",
       "      <td>Plexin (semaphorin receptor)</td>\n",
       "      <td>Is an ortholog of human SLIT2 (slit guidance l...</td>\n",
       "      <td>Is an ortholog of human PLXNC1 (plexin C1) and...</td>\n",
       "      <td>Putative based on literature</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase; \"extracellular region\", https://www....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>WBGene00001980</td>\n",
       "      <td>WBGene00001980</td>\n",
       "      <td>hmr-1</td>\n",
       "      <td>hmr-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>144</td>\n",
       "      <td>Cell adhesion</td>\n",
       "      <td>E-cadherin</td>\n",
       "      <td>E-cadherin</td>\n",
       "      <td>Exhibits beta-catenin binding activity and del...</td>\n",
       "      <td>Exhibits beta-catenin binding activity and del...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>PMCID: PMC4449804</td>\n",
       "      <td>wormbase; \"Located in adherens junction. Is sp...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>WBGene00004264</td>\n",
       "      <td>WBGene00004210</td>\n",
       "      <td>qua-1</td>\n",
       "      <td>ptc-3</td>\n",
       "      <td>1.0</td>\n",
       "      <td>148</td>\n",
       "      <td>Hedgehog signaling</td>\n",
       "      <td>IHH</td>\n",
       "      <td>PTCH2</td>\n",
       "      <td>Is an ortholog of human DHH (desert hedgehog s...</td>\n",
       "      <td>Is an ortholog of human PTCH1 (patched 1) and ...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>DOIs: 10.1186/1471-2164-7-280; 10.1016/j.ydbio...</td>\n",
       "      <td>wormbase; \"extracellular region\",  https://www...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>WBGene00003395</td>\n",
       "      <td>WBGene00003238</td>\n",
       "      <td>mom-2</td>\n",
       "      <td>mig-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>150</td>\n",
       "      <td>Wnt signaling</td>\n",
       "      <td>Wnt</td>\n",
       "      <td>Frizzled</td>\n",
       "      <td>Exhibits receptor tyrosine kinase binding acti...</td>\n",
       "      <td>Is predicted to have Wnt-activated receptor ac...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...</td>\n",
       "      <td>wormbase; \"extracellular region/space\"; https:...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>WBGene00003395</td>\n",
       "      <td>WBGene00003006</td>\n",
       "      <td>mom-2</td>\n",
       "      <td>lin-17</td>\n",
       "      <td>1.0</td>\n",
       "      <td>157</td>\n",
       "      <td>Wnt signaling</td>\n",
       "      <td>Wnt</td>\n",
       "      <td>Frizzled</td>\n",
       "      <td>Exhibits receptor tyrosine kinase binding acti...</td>\n",
       "      <td>Is an ortholog of human FZD4 (frizzled class r...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...</td>\n",
       "      <td>wormbase; \"extracellular region/space\"; https:...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>WBGene00004732</td>\n",
       "      <td>WBGene00002299</td>\n",
       "      <td>sax-7</td>\n",
       "      <td>let-23</td>\n",
       "      <td>1.0</td>\n",
       "      <td>163</td>\n",
       "      <td>Cell adhesion</td>\n",
       "      <td>CAMs</td>\n",
       "      <td>EGF-R</td>\n",
       "      <td>Is an ortholog of human NRCAM (neuronal cell a...</td>\n",
       "      <td>Is an ortholog of human ERBB4 (erb-b2 receptor...</td>\n",
       "      <td>Putative based on literature</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase; \"sax-7...Located in several cellular...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>WBGene00002992</td>\n",
       "      <td>WBGene00002299</td>\n",
       "      <td>lin-3</td>\n",
       "      <td>let-23</td>\n",
       "      <td>1.0</td>\n",
       "      <td>164</td>\n",
       "      <td>Canonical RTK-Ras-ERK signaling</td>\n",
       "      <td>EGF</td>\n",
       "      <td>EGF-R</td>\n",
       "      <td>Is predicted to have epidermal growth factor r...</td>\n",
       "      <td>Is an ortholog of human ERBB4 (erb-b2 receptor...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>WormBook</td>\n",
       "      <td>wormbase; \"Located in plasma membrane\"</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>WBGene00006951</td>\n",
       "      <td>WBGene00004208</td>\n",
       "      <td>wrt-5</td>\n",
       "      <td>ptc-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>168</td>\n",
       "      <td>Hedgehog signaling</td>\n",
       "      <td>DHH</td>\n",
       "      <td>PTCH1</td>\n",
       "      <td>Is predicted to have signaling receptor bindin...</td>\n",
       "      <td>Is an ortholog of human PTCH1 (patched 1) and ...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>DOIs: 10.1186/1471-2164-7-280; 10.1016/j.ydbio...</td>\n",
       "      <td>wormbase; \"Predicted to be located in extracel...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>WBGene00002085</td>\n",
       "      <td>WBGene00000898</td>\n",
       "      <td>ins-2</td>\n",
       "      <td>daf-2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>176</td>\n",
       "      <td>Insulin signaling</td>\n",
       "      <td>ILP</td>\n",
       "      <td>IGFR</td>\n",
       "      <td>Is predicted to have hormone activity. Is invo...</td>\n",
       "      <td>Is an ortholog of human IGF1R (insulin like gr...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>PMCID: PMC6204898</td>\n",
       "      <td>wormbase; \"Predicted to be located in extracel...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>WBGene00002091</td>\n",
       "      <td>WBGene00000898</td>\n",
       "      <td>ins-8</td>\n",
       "      <td>daf-2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>180</td>\n",
       "      <td>Insulin signaling</td>\n",
       "      <td>ILP</td>\n",
       "      <td>IGFR</td>\n",
       "      <td>Is predicted to have hormone activity. Is expr...</td>\n",
       "      <td>Is an ortholog of human IGF1R (insulin like gr...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>PMCID: PMC6204898</td>\n",
       "      <td>wormbase; \"Predicted to be located in extracel...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>WBGene00002112</td>\n",
       "      <td>WBGene00000898</td>\n",
       "      <td>ins-29</td>\n",
       "      <td>daf-2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>198</td>\n",
       "      <td>Insulin signaling</td>\n",
       "      <td>ILP</td>\n",
       "      <td>IGFR</td>\n",
       "      <td>Is predicted to have hormone activity. Is expr...</td>\n",
       "      <td>Is an ortholog of human IGF1R (insulin like gr...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>PMCID: PMC6204898</td>\n",
       "      <td>worm base; disruption of gene impacts longevit...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>WBGene00002114</td>\n",
       "      <td>WBGene00000898</td>\n",
       "      <td>ins-31</td>\n",
       "      <td>daf-2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>200</td>\n",
       "      <td>Insulin signaling</td>\n",
       "      <td>ILP</td>\n",
       "      <td>IGFR</td>\n",
       "      <td>Is predicted to have hormone activity. Is expr...</td>\n",
       "      <td>Is an ortholog of human IGF1R (insulin like gr...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>PMCID: PMC6204898</td>\n",
       "      <td>worm base; disruption of gene impacts longevit...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>WBGene00002117</td>\n",
       "      <td>WBGene00000898</td>\n",
       "      <td>ins-34</td>\n",
       "      <td>daf-2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>203</td>\n",
       "      <td>Insulin signaling</td>\n",
       "      <td>ILP</td>\n",
       "      <td>IGFR</td>\n",
       "      <td>Is expressed in several structures, including ...</td>\n",
       "      <td>Is an ortholog of human IGF1R (insulin like gr...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>PMCID: PMC6204898</td>\n",
       "      <td>worm base; disruption of gene impacts longevit...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>WBGene00002119</td>\n",
       "      <td>WBGene00000898</td>\n",
       "      <td>ins-36</td>\n",
       "      <td>daf-2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>205</td>\n",
       "      <td>Insulin signaling</td>\n",
       "      <td>ILP</td>\n",
       "      <td>IGFR</td>\n",
       "      <td>Is expressed in several structures, including ...</td>\n",
       "      <td>Is an ortholog of human IGF1R (insulin like gr...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>PMCID: PMC6204898</td>\n",
       "      <td>worm base; disruption of gene impacts longevit...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>WBGene00002120</td>\n",
       "      <td>WBGene00000898</td>\n",
       "      <td>ins-37</td>\n",
       "      <td>daf-2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>206</td>\n",
       "      <td>Insulin signaling</td>\n",
       "      <td>ILP</td>\n",
       "      <td>IGFR</td>\n",
       "      <td>Is predicted to have hormone activity. Is expr...</td>\n",
       "      <td>Is an ortholog of human IGF1R (insulin like gr...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Secreted</td>\n",
       "      <td>PMCID: PMC6204898</td>\n",
       "      <td>worm base; disruption of gene impacts longevit...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>WBGene00001104</td>\n",
       "      <td>WBGene00003001</td>\n",
       "      <td>dsl-2</td>\n",
       "      <td>lin-12</td>\n",
       "      <td>1.0</td>\n",
       "      <td>211</td>\n",
       "      <td>Notch signaling</td>\n",
       "      <td>DLL</td>\n",
       "      <td>Notch</td>\n",
       "      <td>Is expressed in head and tail. Is predicted to...</td>\n",
       "      <td>Is an ortholog of human NOTCH4 (notch receptor...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>PMCID: PMC3102077</td>\n",
       "      <td>worm base; component of plasma membrane: https...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>WBGene00000168</td>\n",
       "      <td>WBGene00001609</td>\n",
       "      <td>apx-1</td>\n",
       "      <td>glp-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>217</td>\n",
       "      <td>Notch signaling</td>\n",
       "      <td>DLL</td>\n",
       "      <td>Notch</td>\n",
       "      <td>Is an ortholog of human DLL3 (delta like canon...</td>\n",
       "      <td>Is an ortholog of human NOTCH1 (notch receptor...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>WormBook</td>\n",
       "      <td>wormbase; \"The maternal genes apx-1 and glp-1 ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>WBGene00001103</td>\n",
       "      <td>WBGene00001609</td>\n",
       "      <td>dsl-1</td>\n",
       "      <td>glp-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>219</td>\n",
       "      <td>Notch signaling</td>\n",
       "      <td>DLL</td>\n",
       "      <td>Notch</td>\n",
       "      <td>Exhibits Notch binding activity. Is expressed ...</td>\n",
       "      <td>Is an ortholog of human NOTCH1 (notch receptor...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>WormBook</td>\n",
       "      <td>wormbase; dsl-1 in endoplasmic reticulum: http...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>WBGene00001104</td>\n",
       "      <td>WBGene00001609</td>\n",
       "      <td>dsl-2</td>\n",
       "      <td>glp-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>220</td>\n",
       "      <td>Notch signaling</td>\n",
       "      <td>DLL</td>\n",
       "      <td>Notch</td>\n",
       "      <td>Is expressed in head and tail. Is predicted to...</td>\n",
       "      <td>Is an ortholog of human NOTCH1 (notch receptor...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>PMCID: PMC3102077</td>\n",
       "      <td>worm base; component of plasma membrane: https...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>WBGene00001106</td>\n",
       "      <td>WBGene00001609</td>\n",
       "      <td>dsl-4</td>\n",
       "      <td>glp-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>222</td>\n",
       "      <td>Notch signaling</td>\n",
       "      <td>DLL</td>\n",
       "      <td>Notch</td>\n",
       "      <td>Is enriched in coelomocyte based on tiling arr...</td>\n",
       "      <td>Is an ortholog of human NOTCH1 (notch receptor...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>PMCID: PMC3102077</td>\n",
       "      <td>wormbase; component of plasma membrane:</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>WBGene00001107</td>\n",
       "      <td>WBGene00001609</td>\n",
       "      <td>dsl-5</td>\n",
       "      <td>glp-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>223</td>\n",
       "      <td>Notch signaling</td>\n",
       "      <td>DLL</td>\n",
       "      <td>Notch</td>\n",
       "      <td>Is enriched in Z1.p; Z4.a; gon_male_dtc anteri...</td>\n",
       "      <td>Is an ortholog of human NOTCH1 (notch receptor...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>PMCID: PMC3102077</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>WBGene00001108</td>\n",
       "      <td>WBGene00001609</td>\n",
       "      <td>dsl-6</td>\n",
       "      <td>glp-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>224</td>\n",
       "      <td>Notch signaling</td>\n",
       "      <td>DLL</td>\n",
       "      <td>Notch</td>\n",
       "      <td>Is enriched in Z1.p; Z4.a; gon_male_dtc anteri...</td>\n",
       "      <td>Is an ortholog of human NOTCH1 (notch receptor...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>PMCID: PMC3102077</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>WBGene00019407</td>\n",
       "      <td>WBGene00011345</td>\n",
       "      <td>K05F1.5</td>\n",
       "      <td>dma-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>230</td>\n",
       "      <td>Cell migration</td>\n",
       "      <td>LECT2</td>\n",
       "      <td>Neuronal tran-mb R</td>\n",
       "      <td>Is an ortholog of human LECT2 (leukocyte cell ...</td>\n",
       "      <td>Exhibits protease binding activity. Is involve...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>ECM-component</td>\n",
       "      <td>DOI: 10.7554/eLife.18345</td>\n",
       "      <td>AG</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>WBGene00003774</td>\n",
       "      <td>WBGene00011345</td>\n",
       "      <td>mnr-1</td>\n",
       "      <td>dma-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>232</td>\n",
       "      <td>Cell migration</td>\n",
       "      <td>GRIN1</td>\n",
       "      <td>Neuronal tran-mb R</td>\n",
       "      <td>Is an ortholog of human GRIN1 (glutamate ionot...</td>\n",
       "      <td>Exhibits protease binding activity. Is involve...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>DOI: 10.1016/j.cell.2013.08.059</td>\n",
       "      <td>wormbase; found in both extracellular space an...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>WBGene00003165</td>\n",
       "      <td>WBGene00003174</td>\n",
       "      <td>mec-1</td>\n",
       "      <td>mec-10</td>\n",
       "      <td>1.0</td>\n",
       "      <td>239</td>\n",
       "      <td>Mechanosensory mechanism</td>\n",
       "      <td>ECM protein</td>\n",
       "      <td>DEG/ENaC channel (receptor of ECM mec proteins)</td>\n",
       "      <td>Is an ortholog of human TFPI2 (tissue factor p...</td>\n",
       "      <td>Is an ortholog of human SCNN1A (sodium channel...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>ECM-component</td>\n",
       "      <td>DOI: 10.1186/s12929-018-0448-y</td>\n",
       "      <td>wormbase; detected in bone marrow, blood, lymp...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>WBGene00003169</td>\n",
       "      <td>WBGene00003174</td>\n",
       "      <td>mec-5</td>\n",
       "      <td>mec-10</td>\n",
       "      <td>1.0</td>\n",
       "      <td>240</td>\n",
       "      <td>Mechanosensory mechanism</td>\n",
       "      <td>ECM protein</td>\n",
       "      <td>DEG/ENaC channel (receptor of ECM mec proteins)</td>\n",
       "      <td>Is predicted to have extracellular matrix stru...</td>\n",
       "      <td>Is an ortholog of human SCNN1A (sodium channel...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>ECM-component</td>\n",
       "      <td>DOI: 10.1186/s12929-018-0448-y</td>\n",
       "      <td>wormbase; made by the epidermal cells that sur...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>WBGene00003173</td>\n",
       "      <td>WBGene00003174</td>\n",
       "      <td>mec-9</td>\n",
       "      <td>mec-10</td>\n",
       "      <td>1.0</td>\n",
       "      <td>241</td>\n",
       "      <td>Mechanosensory mechanism</td>\n",
       "      <td>ECM protein</td>\n",
       "      <td>DEG/ENaC channel (receptor of ECM mec proteins)</td>\n",
       "      <td>Is an ortholog of human SPINT2 (serine peptida...</td>\n",
       "      <td>Is an ortholog of human SCNN1A (sodium channel...</td>\n",
       "      <td>Literature supported</td>\n",
       "      <td>ECM-component</td>\n",
       "      <td>DOI: 10.1186/s12929-018-0448-y</td>\n",
       "      <td>wormbase; made by the epidermal cells that sur...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>WBGene00000693</td>\n",
       "      <td>WBGene00000900</td>\n",
       "      <td>col-119</td>\n",
       "      <td>daf-4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>243</td>\n",
       "      <td>TGF-B signaling</td>\n",
       "      <td>Collagen</td>\n",
       "      <td>TGF-B-R</td>\n",
       "      <td>Is predicted to be a structural constituent of...</td>\n",
       "      <td>Is an ortholog of human ACVR2B (activin A rece...</td>\n",
       "      <td>Putative based on data</td>\n",
       "      <td>Membrane-bound</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase; structural constituent of cuticle: h...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>WBGene00000608</td>\n",
       "      <td>WBGene00000900</td>\n",
       "      <td>col-19</td>\n",
       "      <td>daf-4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>244</td>\n",
       "      <td>TGF-B signaling</td>\n",
       "      <td>Collagen</td>\n",
       "      <td>TGF-B-R</td>\n",
       "      <td>Is a structural constituent of collagen and cu...</td>\n",
       "      <td>Is an ortholog of human ACVR2B (activin A rece...</td>\n",
       "      <td>Putative based on data</td>\n",
       "      <td>ECM-component</td>\n",
       "      <td>Orthology, protein-protein interaction (STRING...</td>\n",
       "      <td>wormbase; ocated in alae of collagen and cutic...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Ligand_WB     Receptor_WB Ligand_symbol Receptor_symbol  score  \\\n",
       "0   WBGene00002881  WBGene00006896       let-756           ver-3    1.0   \n",
       "1   WBGene00002881  WBGene00006894       let-756           ver-1    1.0   \n",
       "2   WBGene00003029  WBGene00003238        lin-44           mig-1    1.0   \n",
       "3   WBGene00002280  WBGene00003930         let-2           pat-3    1.0   \n",
       "4   WBGene00001164  WBGene00006868         efn-3           vab-1    1.0   \n",
       "5   WBGene00000674  WBGene00016104        col-99           ddr-1    1.0   \n",
       "6   WBGene00003395  WBGene00003397         mom-2           mom-5    1.0   \n",
       "7   WBGene00006746  WBGene00007764         unc-6         C27B7.7    1.0   \n",
       "8   WBGene00001700  WBGene00004208        grd-11           ptc-1    1.0   \n",
       "9   WBGene00000857  WBGene00003238         cwn-1           mig-1    1.0   \n",
       "10  WBGene00006750  WBGene00006765        unc-10          unc-29    1.0   \n",
       "11  WBGene00006926  WBGene00004374         vit-2           rme-2    1.0   \n",
       "12  WBGene00002248  WBGene00003930         lam-3           pat-3    1.0   \n",
       "13  WBGene00001188  WBGene00003238        egl-20           mig-1    1.0   \n",
       "14  WBGene00001103  WBGene00003001         dsl-1          lin-12    1.0   \n",
       "15  WBGene00003111  WBGene00004048        mab-20           plx-2    1.0   \n",
       "16  WBGene00003029  WBGene00000289        lin-44           cam-1    1.0   \n",
       "17  WBGene00021594  WBGene00000900         tig-3           daf-4    1.0   \n",
       "18  WBGene00003395  WBGene00003007         mom-2          lin-18    1.0   \n",
       "19  WBGene00000858  WBGene00000478         cwn-2           cfz-2    1.0   \n",
       "20  WBGene00003029  WBGene00000478        lin-44           cfz-2    1.0   \n",
       "21  WBGene00002280  WBGene00001184         let-2          egl-15    1.0   \n",
       "22  WBGene00004889  WBGene00004048         smp-1           plx-2    1.0   \n",
       "23  WBGene00004890  WBGene00004048         smp-2           plx-2    1.0   \n",
       "24  WBGene00004854  WBGene00004048         slt-1           plx-2    1.0   \n",
       "25  WBGene00001980  WBGene00001980         hmr-1           hmr-1    1.0   \n",
       "26  WBGene00004264  WBGene00004210         qua-1           ptc-3    1.0   \n",
       "27  WBGene00003395  WBGene00003238         mom-2           mig-1    1.0   \n",
       "28  WBGene00003395  WBGene00003006         mom-2          lin-17    1.0   \n",
       "29  WBGene00004732  WBGene00002299         sax-7          let-23    1.0   \n",
       "30  WBGene00002992  WBGene00002299         lin-3          let-23    1.0   \n",
       "31  WBGene00006951  WBGene00004208         wrt-5           ptc-1    1.0   \n",
       "32  WBGene00002085  WBGene00000898         ins-2           daf-2    1.0   \n",
       "33  WBGene00002091  WBGene00000898         ins-8           daf-2    1.0   \n",
       "34  WBGene00002112  WBGene00000898        ins-29           daf-2    1.0   \n",
       "35  WBGene00002114  WBGene00000898        ins-31           daf-2    1.0   \n",
       "36  WBGene00002117  WBGene00000898        ins-34           daf-2    1.0   \n",
       "37  WBGene00002119  WBGene00000898        ins-36           daf-2    1.0   \n",
       "38  WBGene00002120  WBGene00000898        ins-37           daf-2    1.0   \n",
       "39  WBGene00001104  WBGene00003001         dsl-2          lin-12    1.0   \n",
       "40  WBGene00000168  WBGene00001609         apx-1           glp-1    1.0   \n",
       "41  WBGene00001103  WBGene00001609         dsl-1           glp-1    1.0   \n",
       "42  WBGene00001104  WBGene00001609         dsl-2           glp-1    1.0   \n",
       "43  WBGene00001106  WBGene00001609         dsl-4           glp-1    1.0   \n",
       "44  WBGene00001107  WBGene00001609         dsl-5           glp-1    1.0   \n",
       "45  WBGene00001108  WBGene00001609         dsl-6           glp-1    1.0   \n",
       "46  WBGene00019407  WBGene00011345       K05F1.5           dma-1    1.0   \n",
       "47  WBGene00003774  WBGene00011345         mnr-1           dma-1    1.0   \n",
       "48  WBGene00003165  WBGene00003174         mec-1          mec-10    1.0   \n",
       "49  WBGene00003169  WBGene00003174         mec-5          mec-10    1.0   \n",
       "50  WBGene00003173  WBGene00003174         mec-9          mec-10    1.0   \n",
       "51  WBGene00000693  WBGene00000900       col-119           daf-4    1.0   \n",
       "52  WBGene00000608  WBGene00000900        col-19           daf-4    1.0   \n",
       "\n",
       "    index                      LR Function                         L Function  \\\n",
       "0       6  Canonical RTK-Ras-ERK signaling                                FGF   \n",
       "1       7  Canonical RTK-Ras-ERK signaling                                FGF   \n",
       "2      25                    Wnt signaling                                Wnt   \n",
       "3      26                    Cell adhesion                           Collagen   \n",
       "4      27           Eph receptor signaling                             Ephrin   \n",
       "5      35                    Cell adhesion                           Collagen   \n",
       "6      38                    Wnt signaling                                Wnt   \n",
       "7      44                   Cell migration            Netrin (dorsal-ventral)   \n",
       "8      50               Hedgehog signaling                                DHH   \n",
       "9      60                    Wnt signaling                                Wnt   \n",
       "10     62           Neuronal communication  Organization of presynaptic sites   \n",
       "11     63    Receptor-Mediated-Endocytosis                       rme-2 ligand   \n",
       "12     64                    Cell adhesion                            Laminin   \n",
       "13     76                    Wnt signaling                                Wnt   \n",
       "14     81                  Notch signaling                              Notch   \n",
       "15     86                   Cell migration         Semaphorin (axon guidance)   \n",
       "16     95                    Wnt signaling                                Wnt   \n",
       "17     97                  TGF-B signaling                                BMP   \n",
       "18    101                    Wnt signaling                                Wnt   \n",
       "19    111                    Wnt signaling                                Wnt   \n",
       "20    112                    Wnt signaling                                Wnt   \n",
       "21    131                    Cell adhesion                           Collagen   \n",
       "22    134                   Cell migration                         Semaphorin   \n",
       "23    135                   Cell migration                         Semaphorin   \n",
       "24    136                   Cell migration           Slit (dorsal nerve cord)   \n",
       "25    144                    Cell adhesion                         E-cadherin   \n",
       "26    148               Hedgehog signaling                                IHH   \n",
       "27    150                    Wnt signaling                                Wnt   \n",
       "28    157                    Wnt signaling                                Wnt   \n",
       "29    163                    Cell adhesion                               CAMs   \n",
       "30    164  Canonical RTK-Ras-ERK signaling                                EGF   \n",
       "31    168               Hedgehog signaling                                DHH   \n",
       "32    176                Insulin signaling                                ILP   \n",
       "33    180                Insulin signaling                                ILP   \n",
       "34    198                Insulin signaling                                ILP   \n",
       "35    200                Insulin signaling                                ILP   \n",
       "36    203                Insulin signaling                                ILP   \n",
       "37    205                Insulin signaling                                ILP   \n",
       "38    206                Insulin signaling                                ILP   \n",
       "39    211                  Notch signaling                                DLL   \n",
       "40    217                  Notch signaling                                DLL   \n",
       "41    219                  Notch signaling                                DLL   \n",
       "42    220                  Notch signaling                                DLL   \n",
       "43    222                  Notch signaling                                DLL   \n",
       "44    223                  Notch signaling                                DLL   \n",
       "45    224                  Notch signaling                                DLL   \n",
       "46    230                   Cell migration                              LECT2   \n",
       "47    232                   Cell migration                              GRIN1   \n",
       "48    239         Mechanosensory mechanism                        ECM protein   \n",
       "49    240         Mechanosensory mechanism                        ECM protein   \n",
       "50    241         Mechanosensory mechanism                        ECM protein   \n",
       "51    243                  TGF-B signaling                           Collagen   \n",
       "52    244                  TGF-B signaling                           Collagen   \n",
       "\n",
       "                                         R Function  \\\n",
       "0             Tyrosine-protein kinase (GF-Receptor)   \n",
       "1                     fms-related tyrosine kinase 1   \n",
       "2                                          Frizzled   \n",
       "3                                          Integrin   \n",
       "4                                              EphR   \n",
       "5                         Discoidin domain receptor   \n",
       "6                                          Frizzled   \n",
       "7                              Ig domain - Receptor   \n",
       "8                                             PTCH1   \n",
       "9                                          Frizzled   \n",
       "10                             Cholinergic Receptor   \n",
       "11                                     LDL receptor   \n",
       "12                                         Integrin   \n",
       "13                                         Frizzled   \n",
       "14                                            Notch   \n",
       "15                     Plexin (semaphorin receptor)   \n",
       "16             Wnt-Receptor (Antagonistic activity)   \n",
       "17                                          TGF-B-R   \n",
       "18                                     Ryk/Derailed   \n",
       "19                                         Frizzled   \n",
       "20                                         Frizzled   \n",
       "21                                            FGF-R   \n",
       "22                     Plexin (semaphorin receptor)   \n",
       "23                     Plexin (semaphorin receptor)   \n",
       "24                     Plexin (semaphorin receptor)   \n",
       "25                                       E-cadherin   \n",
       "26                                            PTCH2   \n",
       "27                                         Frizzled   \n",
       "28                                         Frizzled   \n",
       "29                                            EGF-R   \n",
       "30                                            EGF-R   \n",
       "31                                            PTCH1   \n",
       "32                                             IGFR   \n",
       "33                                             IGFR   \n",
       "34                                             IGFR   \n",
       "35                                             IGFR   \n",
       "36                                             IGFR   \n",
       "37                                             IGFR   \n",
       "38                                             IGFR   \n",
       "39                                            Notch   \n",
       "40                                            Notch   \n",
       "41                                            Notch   \n",
       "42                                            Notch   \n",
       "43                                            Notch   \n",
       "44                                            Notch   \n",
       "45                                            Notch   \n",
       "46                               Neuronal tran-mb R   \n",
       "47                               Neuronal tran-mb R   \n",
       "48  DEG/ENaC channel (receptor of ECM mec proteins)   \n",
       "49  DEG/ENaC channel (receptor of ECM mec proteins)   \n",
       "50  DEG/ENaC channel (receptor of ECM mec proteins)   \n",
       "51                                          TGF-B-R   \n",
       "52                                          TGF-B-R   \n",
       "\n",
       "                    Ligand_desc_WormBase_versionWS277  \\\n",
       "0   Is an ortholog of human FGF16 (fibroblast grow...   \n",
       "1   Is an ortholog of human FGF16 (fibroblast grow...   \n",
       "2   Is an ortholog of human WNT10A (Wnt family mem...   \n",
       "3   Is an ortholog of human COL4A5 (collagen type ...   \n",
       "4   Is an ortholog of human EFNB1 (ephrin B1); EFN...   \n",
       "5   Is an ortholog of human COL13A1 (collagen type...   \n",
       "6   Exhibits receptor tyrosine kinase binding acti...   \n",
       "7   Is an ortholog of human NTN1 (netrin 1). Is pr...   \n",
       "8   Is an ortholog of human DHH (desert hedgehog s...   \n",
       "9   Is an ortholog of human WNT4 (Wnt family membe...   \n",
       "10  Is an ortholog of human RIMS1 (regulating syna...   \n",
       "11  Is predicted to have lipid transporter activit...   \n",
       "12  Is an ortholog of human LAMA2 (laminin subunit...   \n",
       "13  Is an ortholog of human WNT16 (Wnt family memb...   \n",
       "14  Exhibits Notch binding activity. Is expressed ...   \n",
       "15  Is an ortholog of human SEMA4D (semaphorin 4D)...   \n",
       "16  Is an ortholog of human WNT10A (Wnt family mem...   \n",
       "17  Is predicted to have cytokine activity. Human ...   \n",
       "18  Exhibits receptor tyrosine kinase binding acti...   \n",
       "19  Is an ortholog of human WNT5A (Wnt family memb...   \n",
       "20  Is an ortholog of human WNT10A (Wnt family mem...   \n",
       "21  Is an ortholog of human COL4A5 (collagen type ...   \n",
       "22  Is an ortholog of human SEMA6A (semaphorin 6A)...   \n",
       "23  Is an ortholog of several human genes includin...   \n",
       "24  Is an ortholog of human SLIT2 (slit guidance l...   \n",
       "25  Exhibits beta-catenin binding activity and del...   \n",
       "26  Is an ortholog of human DHH (desert hedgehog s...   \n",
       "27  Exhibits receptor tyrosine kinase binding acti...   \n",
       "28  Exhibits receptor tyrosine kinase binding acti...   \n",
       "29  Is an ortholog of human NRCAM (neuronal cell a...   \n",
       "30  Is predicted to have epidermal growth factor r...   \n",
       "31  Is predicted to have signaling receptor bindin...   \n",
       "32  Is predicted to have hormone activity. Is invo...   \n",
       "33  Is predicted to have hormone activity. Is expr...   \n",
       "34  Is predicted to have hormone activity. Is expr...   \n",
       "35  Is predicted to have hormone activity. Is expr...   \n",
       "36  Is expressed in several structures, including ...   \n",
       "37  Is expressed in several structures, including ...   \n",
       "38  Is predicted to have hormone activity. Is expr...   \n",
       "39  Is expressed in head and tail. Is predicted to...   \n",
       "40  Is an ortholog of human DLL3 (delta like canon...   \n",
       "41  Exhibits Notch binding activity. Is expressed ...   \n",
       "42  Is expressed in head and tail. Is predicted to...   \n",
       "43  Is enriched in coelomocyte based on tiling arr...   \n",
       "44  Is enriched in Z1.p; Z4.a; gon_male_dtc anteri...   \n",
       "45  Is enriched in Z1.p; Z4.a; gon_male_dtc anteri...   \n",
       "46  Is an ortholog of human LECT2 (leukocyte cell ...   \n",
       "47  Is an ortholog of human GRIN1 (glutamate ionot...   \n",
       "48  Is an ortholog of human TFPI2 (tissue factor p...   \n",
       "49  Is predicted to have extracellular matrix stru...   \n",
       "50  Is an ortholog of human SPINT2 (serine peptida...   \n",
       "51  Is predicted to be a structural constituent of...   \n",
       "52  Is a structural constituent of collagen and cu...   \n",
       "\n",
       "                  Receptor_desc_WormBase_versionWS277  \\\n",
       "0   Is an ortholog of human FLT1 (fms related rece...   \n",
       "1   Is an ortholog of human FLT1 (fms related rece...   \n",
       "2   Is predicted to have Wnt-activated receptor ac...   \n",
       "3   Is an ortholog of human ITGB1 (integrin subuni...   \n",
       "4   Is an ortholog of human EPHA1 (EPH receptor A1...   \n",
       "5   Is an ortholog of human DDR1 (discoidin domain...   \n",
       "6   Is an ortholog of human FZD1 (frizzled class r...   \n",
       "7   Is an ortholog of human DSCAM (DS cell adhesio...   \n",
       "8   Is an ortholog of human PTCH1 (patched 1) and ...   \n",
       "9   Is predicted to have Wnt-activated receptor ac...   \n",
       "10  Is an ortholog of several human genes includin...   \n",
       "11  Exhibits low-density lipoprotein particle rece...   \n",
       "12  Is an ortholog of human ITGB1 (integrin subuni...   \n",
       "13  Is predicted to have Wnt-activated receptor ac...   \n",
       "14  Is an ortholog of human NOTCH4 (notch receptor...   \n",
       "15  Is an ortholog of human PLXNC1 (plexin C1) and...   \n",
       "16  Is an ortholog of human ROR1 (receptor tyrosin...   \n",
       "17  Is an ortholog of human ACVR2B (activin A rece...   \n",
       "18  Is an ortholog of human RYK (receptor like tyr...   \n",
       "19  Is an ortholog of human FZD5 (frizzled class r...   \n",
       "20  Is an ortholog of human FZD5 (frizzled class r...   \n",
       "21  Is an ortholog of human FGFR1 (fibroblast grow...   \n",
       "22  Is an ortholog of human PLXNC1 (plexin C1) and...   \n",
       "23  Is an ortholog of human PLXNC1 (plexin C1) and...   \n",
       "24  Is an ortholog of human PLXNC1 (plexin C1) and...   \n",
       "25  Exhibits beta-catenin binding activity and del...   \n",
       "26  Is an ortholog of human PTCH1 (patched 1) and ...   \n",
       "27  Is predicted to have Wnt-activated receptor ac...   \n",
       "28  Is an ortholog of human FZD4 (frizzled class r...   \n",
       "29  Is an ortholog of human ERBB4 (erb-b2 receptor...   \n",
       "30  Is an ortholog of human ERBB4 (erb-b2 receptor...   \n",
       "31  Is an ortholog of human PTCH1 (patched 1) and ...   \n",
       "32  Is an ortholog of human IGF1R (insulin like gr...   \n",
       "33  Is an ortholog of human IGF1R (insulin like gr...   \n",
       "34  Is an ortholog of human IGF1R (insulin like gr...   \n",
       "35  Is an ortholog of human IGF1R (insulin like gr...   \n",
       "36  Is an ortholog of human IGF1R (insulin like gr...   \n",
       "37  Is an ortholog of human IGF1R (insulin like gr...   \n",
       "38  Is an ortholog of human IGF1R (insulin like gr...   \n",
       "39  Is an ortholog of human NOTCH4 (notch receptor...   \n",
       "40  Is an ortholog of human NOTCH1 (notch receptor...   \n",
       "41  Is an ortholog of human NOTCH1 (notch receptor...   \n",
       "42  Is an ortholog of human NOTCH1 (notch receptor...   \n",
       "43  Is an ortholog of human NOTCH1 (notch receptor...   \n",
       "44  Is an ortholog of human NOTCH1 (notch receptor...   \n",
       "45  Is an ortholog of human NOTCH1 (notch receptor...   \n",
       "46  Exhibits protease binding activity. Is involve...   \n",
       "47  Exhibits protease binding activity. Is involve...   \n",
       "48  Is an ortholog of human SCNN1A (sodium channel...   \n",
       "49  Is an ortholog of human SCNN1A (sodium channel...   \n",
       "50  Is an ortholog of human SCNN1A (sodium channel...   \n",
       "51  Is an ortholog of human ACVR2B (activin A rece...   \n",
       "52  Is an ortholog of human ACVR2B (activin A rece...   \n",
       "\n",
       "                 Pair_confidence   Location_type  \\\n",
       "0   Putative based on literature        Secreted   \n",
       "1   Putative based on literature        Secreted   \n",
       "2           Literature supported        Secreted   \n",
       "3           Literature supported   ECM-component   \n",
       "4           Literature supported  Membrane-bound   \n",
       "5           Literature supported   ECM-component   \n",
       "6           Literature supported        Secreted   \n",
       "7   Putative based on literature        Secreted   \n",
       "8           Literature supported  Membrane-bound   \n",
       "9           Literature supported        Secreted   \n",
       "10  Putative based on literature  Membrane-bound   \n",
       "11          Literature supported        Secreted   \n",
       "12          Literature supported   ECM-component   \n",
       "13          Literature supported        Secreted   \n",
       "14          Literature supported  Membrane-bound   \n",
       "15          Literature supported  Membrane-bound   \n",
       "16          Literature supported        Secreted   \n",
       "17          Literature supported        Secreted   \n",
       "18          Literature supported        Secreted   \n",
       "19          Literature supported        Secreted   \n",
       "20          Literature supported        Secreted   \n",
       "21  Putative based on literature   ECM-component   \n",
       "22  Putative based on literature        Secreted   \n",
       "23  Putative based on literature        Secreted   \n",
       "24  Putative based on literature        Secreted   \n",
       "25          Literature supported  Membrane-bound   \n",
       "26          Literature supported        Secreted   \n",
       "27          Literature supported        Secreted   \n",
       "28          Literature supported        Secreted   \n",
       "29  Putative based on literature  Membrane-bound   \n",
       "30          Literature supported  Membrane-bound   \n",
       "31          Literature supported        Secreted   \n",
       "32          Literature supported        Secreted   \n",
       "33          Literature supported        Secreted   \n",
       "34          Literature supported        Secreted   \n",
       "35          Literature supported        Secreted   \n",
       "36          Literature supported        Secreted   \n",
       "37          Literature supported        Secreted   \n",
       "38          Literature supported        Secreted   \n",
       "39          Literature supported  Membrane-bound   \n",
       "40          Literature supported  Membrane-bound   \n",
       "41          Literature supported  Membrane-bound   \n",
       "42          Literature supported  Membrane-bound   \n",
       "43          Literature supported  Membrane-bound   \n",
       "44          Literature supported  Membrane-bound   \n",
       "45          Literature supported  Membrane-bound   \n",
       "46          Literature supported   ECM-component   \n",
       "47          Literature supported  Membrane-bound   \n",
       "48          Literature supported   ECM-component   \n",
       "49          Literature supported   ECM-component   \n",
       "50          Literature supported   ECM-component   \n",
       "51        Putative based on data  Membrane-bound   \n",
       "52        Putative based on data   ECM-component   \n",
       "\n",
       "                                  Confidence_evidence  \\\n",
       "0   Orthology, protein-protein interaction (STRING...   \n",
       "1   Orthology, protein-protein interaction (STRING...   \n",
       "2   WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...   \n",
       "3                               10.1083/jcb.201903124   \n",
       "4                                            WormBook   \n",
       "5                      DOI: 10.1016/j.mcn.2018.03.003   \n",
       "6   WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...   \n",
       "7   Orthology, protein-protein interaction (STRING...   \n",
       "8   DOIs: 10.1186/1471-2164-7-280; 10.1016/j.ydbio...   \n",
       "9   WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...   \n",
       "10  Orthology, protein-protein interaction (STRING...   \n",
       "11                                           WormBook   \n",
       "12                                           WormBook   \n",
       "13  WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...   \n",
       "14                                           WormBook   \n",
       "15                  DOI: 10.1371/journal.pgen.1005950   \n",
       "16                            DOI: 10.1242/dev.005363   \n",
       "17                    DOI: 10.1016/j.isci.2022.103791   \n",
       "18  WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...   \n",
       "19  WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...   \n",
       "20  WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...   \n",
       "21  Orthology, protein-protein interaction (STRING...   \n",
       "22  Orthology, protein-protein interaction (STRING...   \n",
       "23  Orthology, protein-protein interaction (STRING...   \n",
       "24  Orthology, protein-protein interaction (STRING...   \n",
       "25                                  PMCID: PMC4449804   \n",
       "26  DOIs: 10.1186/1471-2164-7-280; 10.1016/j.ydbio...   \n",
       "27  WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...   \n",
       "28  WormBook, DOIs: 10.1016/j.devcel.2006.02.010; ...   \n",
       "29  Orthology, protein-protein interaction (STRING...   \n",
       "30                                           WormBook   \n",
       "31  DOIs: 10.1186/1471-2164-7-280; 10.1016/j.ydbio...   \n",
       "32                                  PMCID: PMC6204898   \n",
       "33                                  PMCID: PMC6204898   \n",
       "34                                  PMCID: PMC6204898   \n",
       "35                                  PMCID: PMC6204898   \n",
       "36                                  PMCID: PMC6204898   \n",
       "37                                  PMCID: PMC6204898   \n",
       "38                                  PMCID: PMC6204898   \n",
       "39                                  PMCID: PMC3102077   \n",
       "40                                           WormBook   \n",
       "41                                           WormBook   \n",
       "42                                  PMCID: PMC3102077   \n",
       "43                                  PMCID: PMC3102077   \n",
       "44                                  PMCID: PMC3102077   \n",
       "45                                  PMCID: PMC3102077   \n",
       "46                           DOI: 10.7554/eLife.18345   \n",
       "47                    DOI: 10.1016/j.cell.2013.08.059   \n",
       "48                     DOI: 10.1186/s12929-018-0448-y   \n",
       "49                     DOI: 10.1186/s12929-018-0448-y   \n",
       "50                     DOI: 10.1186/s12929-018-0448-y   \n",
       "51  Orthology, protein-protein interaction (STRING...   \n",
       "52  Orthology, protein-protein interaction (STRING...   \n",
       "\n",
       "                                    Location_evidence  \n",
       "0   wormbase: let-756 is found in the cytoplasm;ex...  \n",
       "1   wormbase: let-756 is found in the cytoplasm;ex...  \n",
       "2   wormbase: lin-44 is said to be located in the ...  \n",
       "3   Wormbase: let-2 is located in the extracellula...  \n",
       "4   wormbase: efn-3 is located in the plasma membr...  \n",
       "5   wormbase: col-99 is located in the extracellul...  \n",
       "6   https://www.ncbi.nlm.nih.gov/gene?Db=gene&Cmd=...  \n",
       "7   wormbase: unc-6 is found to be located in axon...  \n",
       "8   wormbase: ptc-1 is a receptor that is membrane...  \n",
       "9   wormbase: annotated as found in the extracellu...  \n",
       "10  wormbase: annotated as membrane-bound, not fou...  \n",
       "11  wormbase: located in cytoplasmic vesicle, veis...  \n",
       "12  wormbase: located in basement membrane; https:...  \n",
       "13  wormbase: located in extracellular space; enab...  \n",
       "14  wormbase: predicted to be integral component o...  \n",
       "15  wormbase: located in cell surface; enables sem...  \n",
       "16  wormbase: predicted to be located in extracell...  \n",
       "17  wormbase: predicted to be located in extracell...  \n",
       "18  wormbase: predicted to be located in extracell...  \n",
       "19  wormbase: predicted to be located in extracell...  \n",
       "20  wormbase: predicted to be located in extracell...  \n",
       "21  wormbase; \"An extracellular matrix structural ...  \n",
       "22  wormbase, \"Predicted to be located in extracel...  \n",
       "23  wormbase, \"exracellular region,\" https://pubme...  \n",
       "24  wormbase; \"extracellular region\", https://www....  \n",
       "25  wormbase; \"Located in adherens junction. Is sp...  \n",
       "26  wormbase; \"extracellular region\",  https://www...  \n",
       "27  wormbase; \"extracellular region/space\"; https:...  \n",
       "28  wormbase; \"extracellular region/space\"; https:...  \n",
       "29  wormbase; \"sax-7...Located in several cellular...  \n",
       "30             wormbase; \"Located in plasma membrane\"  \n",
       "31  wormbase; \"Predicted to be located in extracel...  \n",
       "32  wormbase; \"Predicted to be located in extracel...  \n",
       "33  wormbase; \"Predicted to be located in extracel...  \n",
       "34  worm base; disruption of gene impacts longevit...  \n",
       "35  worm base; disruption of gene impacts longevit...  \n",
       "36  worm base; disruption of gene impacts longevit...  \n",
       "37  worm base; disruption of gene impacts longevit...  \n",
       "38  worm base; disruption of gene impacts longevit...  \n",
       "39  worm base; component of plasma membrane: https...  \n",
       "40  wormbase; \"The maternal genes apx-1 and glp-1 ...  \n",
       "41  wormbase; dsl-1 in endoplasmic reticulum: http...  \n",
       "42  worm base; component of plasma membrane: https...  \n",
       "43           wormbase; component of plasma membrane:   \n",
       "44                                                NaN  \n",
       "45                                                NaN  \n",
       "46                                                 AG  \n",
       "47  wormbase; found in both extracellular space an...  \n",
       "48  wormbase; detected in bone marrow, blood, lymp...  \n",
       "49  wormbase; made by the epidermal cells that sur...  \n",
       "50  wormbase; made by the epidermal cells that sur...  \n",
       "51  wormbase; structural constituent of cuticle: h...  \n",
       "52  wormbase; ocated in alae of collagen and cutic...  "
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ga_export"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(53, 15)"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ga_export.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "ga_export.to_csv(data_folder + '/PPI-Networks/Celegans-GA-ICELLNET-Selected-LR-pairs.csv', index=False)\n",
    "ga_export.to_csv(output_folder + '/Celegans-GA-ICELLNET-Selected-LR-pairs.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analyze functions in lists selected across all GA runs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Generate function metadata**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "functions = symbol_lrs['LR Function'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "function_colors = c2c.plotting.get_colors_from_labels(functions, cmap='tab20')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "func_color_dict = dict()\n",
    "row_cols = []\n",
    "row_functions = []\n",
    "for idx, row in symbol_lrs.iterrows():\n",
    "    name = str(row['Ligand_symbol']) + '_' + str(row['Receptor_symbol'])\n",
    "    func = str(row['LR Function'])\n",
    "    func_color_dict[name] = function_colors[func]\n",
    "    row_cols.append((name, function_colors[func]))\n",
    "    row_functions.append((name, func))\n",
    "row_cols = pd.DataFrame.from_records(row_cols, columns=['LR', 'Function']).set_index('LR')\n",
    "row_functions = pd.DataFrame.from_records(row_functions, columns=['LR', 'Function']).set_index('LR')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "func_mapper = row_functions.to_dict()['Function']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Compute fraction of each function in each GA-selected list**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "func_count = df.T\n",
    "func_count.index = df.columns.map(func_mapper)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "func_count = func_count.reset_index().groupby('index').sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "func_frac = func_count / func_count.sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "sorted_order = func_frac.T.median().sort_values(ascending=False).index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Wnt signaling', 'Cell migration', 'Notch signaling', 'Cell adhesion',\n",
       "       'Insulin signaling', 'Mechanosensory mechanism',\n",
       "       'Canonical RTK-Ras-ERK signaling', 'Hedgehog signaling',\n",
       "       'TGF-B signaling', 'Neuronal communication', 'Eph receptor signaling',\n",
       "       'Receptor-Mediated-Endocytosis', 'Trypsin-like proteolysis',\n",
       "       'Stress response', 'Apoptosis', 'PCP pathway', 'Lipid homeostasis',\n",
       "       'Immune response', 'Epidermal development', 'Longevity'],\n",
       "      dtype='object', name='index')"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sorted_order"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Plot Fraction for each Function**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "composition = func_frac.loc[sorted_order,:].T#.sort_values(by=list(sorted_order)[:5], ascending=False)\n",
    "composition = composition.T\n",
    "composition['Complete'] = symbol_lrs.groupby('LR Function').count()['index'] / len(symbol_lrs)\n",
    "composition[['Complete'] + list(composition.columns)[:-1]].T.plot(kind='bar', stacked=True, cmap='tab20', figsize=(16, 9))\n",
    "plt.legend(loc='center left', bbox_to_anchor=(1.05, 0.5),\n",
    "              ncol=2, fancybox=True, shadow=True, fontsize=26)\n",
    "plt.ylim((0,1))\n",
    "\n",
    "plt.xlabel('List of Ligand-Receptor Pairs from each Run\\nof the Genetic Algorithm', fontsize=30)\n",
    "plt.ylabel('Fraction of Ligand-Receptor Pairs\\n per Function', fontsize=30)\n",
    "\n",
    "plt.tick_params(axis='y', labelsize=28)\n",
    "plt.xticks([])\n",
    "\n",
    "plt.savefig(output_folder + '/Functional-Composition-GA-ICELLNET-Runs.svg',\n",
    "                dpi=300,\n",
    "                bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Presence in all GA runs**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "index\n",
       "Apoptosis                          False\n",
       "Canonical RTK-Ras-ERK signaling    False\n",
       "Cell adhesion                       True\n",
       "Cell migration                      True\n",
       "Eph receptor signaling             False\n",
       "Epidermal development              False\n",
       "Hedgehog signaling                 False\n",
       "Immune response                    False\n",
       "Insulin signaling                  False\n",
       "Lipid homeostasis                  False\n",
       "Longevity                          False\n",
       "Mechanosensory mechanism            True\n",
       "Neuronal communication             False\n",
       "Notch signaling                     True\n",
       "PCP pathway                        False\n",
       "Receptor-Mediated-Endocytosis      False\n",
       "Stress response                    False\n",
       "TGF-B signaling                    False\n",
       "Trypsin-like proteolysis           False\n",
       "Wnt signaling                       True\n",
       "dtype: bool"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(func_frac.T != 0).all()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Presence in at least one GA run**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "index\n",
       "Apoptosis                          False\n",
       "Canonical RTK-Ras-ERK signaling     True\n",
       "Cell adhesion                       True\n",
       "Cell migration                      True\n",
       "Eph receptor signaling              True\n",
       "Epidermal development              False\n",
       "Hedgehog signaling                  True\n",
       "Immune response                     True\n",
       "Insulin signaling                   True\n",
       "Lipid homeostasis                  False\n",
       "Longevity                           True\n",
       "Mechanosensory mechanism            True\n",
       "Neuronal communication              True\n",
       "Notch signaling                     True\n",
       "PCP pathway                        False\n",
       "Receptor-Mediated-Endocytosis       True\n",
       "Stress response                     True\n",
       "TGF-B signaling                     True\n",
       "Trypsin-like proteolysis            True\n",
       "Wnt signaling                       True\n",
       "dtype: bool"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(func_frac.T != 0).any()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Wilcoxon test to detect functions that change proportions across GA runs**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "wilcoxon_functions = pd.DataFrame(columns=['P-val'])\n",
    "for idx, row in func_frac.iterrows():\n",
    "    wilcoxon_functions.at[idx, 'Greater P-val'] = scipy.stats.wilcoxon(row.values-composition.loc[idx, 'Complete'],\n",
    "                                                                       alternative='greater')[1]\n",
    "    wilcoxon_functions.at[idx, 'Less P-val'] = scipy.stats.wilcoxon(row.values-composition.loc[idx, 'Complete'],\n",
    "                                                                       alternative='less')[1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**FDR correction**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "from statsmodels.stats.multitest import multipletests"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "wilcoxon_functions['Greater Q-val'] = multipletests(wilcoxon_functions['Greater P-val'].values.flatten(), alpha=0.01, method='fdr_bh')[1]\n",
    "wilcoxon_functions['Less Q-val'] = multipletests(wilcoxon_functions['Less P-val'].values.flatten(), alpha=0.01, method='fdr_bh')[1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Functions that increased their fraction across the GA-selected lists**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Apoptosis                          False\n",
       "Canonical RTK-Ras-ERK signaling     True\n",
       "Cell adhesion                      False\n",
       "Cell migration                      True\n",
       "Eph receptor signaling             False\n",
       "Epidermal development              False\n",
       "Hedgehog signaling                  True\n",
       "Immune response                    False\n",
       "Insulin signaling                  False\n",
       "Lipid homeostasis                  False\n",
       "Longevity                          False\n",
       "Mechanosensory mechanism            True\n",
       "Neuronal communication              True\n",
       "Notch signaling                     True\n",
       "PCP pathway                        False\n",
       "Receptor-Mediated-Endocytosis      False\n",
       "Stress response                    False\n",
       "TGF-B signaling                    False\n",
       "Trypsin-like proteolysis           False\n",
       "Wnt signaling                      False\n",
       "Name: Greater Q-val, dtype: bool"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wilcoxon_functions['Greater Q-val'] < 0.01"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Functions that decreased their fraction across the GA-selected lists**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Apoptosis                           True\n",
       "Canonical RTK-Ras-ERK signaling    False\n",
       "Cell adhesion                       True\n",
       "Cell migration                     False\n",
       "Eph receptor signaling             False\n",
       "Epidermal development               True\n",
       "Hedgehog signaling                 False\n",
       "Immune response                     True\n",
       "Insulin signaling                   True\n",
       "Lipid homeostasis                   True\n",
       "Longevity                           True\n",
       "Mechanosensory mechanism           False\n",
       "Neuronal communication             False\n",
       "Notch signaling                    False\n",
       "PCP pathway                         True\n",
       "Receptor-Mediated-Endocytosis      False\n",
       "Stress response                     True\n",
       "TGF-B signaling                     True\n",
       "Trypsin-like proteolysis            True\n",
       "Wnt signaling                      False\n",
       "Name: Less Q-val, dtype: bool"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wilcoxon_functions['Less Q-val'] < 0.01"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Functions that neither increased or decreased their fraction in the GA-selected lists**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Apoptosis                          False\n",
       "Canonical RTK-Ras-ERK signaling    False\n",
       "Cell adhesion                      False\n",
       "Cell migration                     False\n",
       "Eph receptor signaling              True\n",
       "Epidermal development              False\n",
       "Hedgehog signaling                 False\n",
       "Immune response                    False\n",
       "Insulin signaling                  False\n",
       "Lipid homeostasis                  False\n",
       "Longevity                          False\n",
       "Mechanosensory mechanism           False\n",
       "Neuronal communication             False\n",
       "Notch signaling                    False\n",
       "PCP pathway                        False\n",
       "Receptor-Mediated-Endocytosis       True\n",
       "Stress response                    False\n",
       "TGF-B signaling                    False\n",
       "Trypsin-like proteolysis           False\n",
       "Wnt signaling                       True\n",
       "dtype: bool"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "~((wilcoxon_functions['Greater Q-val'] < 0.01) | (wilcoxon_functions['Less Q-val'] < 0.01))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "index\n",
       "Wnt signaling                      6.013570e-01\n",
       "Cell migration                     1.825332e-08\n",
       "Notch signaling                    3.038427e-10\n",
       "Cell adhesion                      9.061083e-17\n",
       "Insulin signaling                  1.513428e-10\n",
       "Mechanosensory mechanism           3.787682e-17\n",
       "Canonical RTK-Ras-ERK signaling    4.984380e-13\n",
       "Hedgehog signaling                 5.554761e-16\n",
       "TGF-B signaling                    3.006341e-11\n",
       "Neuronal communication             1.825332e-08\n",
       "Eph receptor signaling             2.337827e-02\n",
       "Receptor-Mediated-Endocytosis      2.807232e-01\n",
       "Trypsin-like proteolysis           3.024558e-03\n",
       "Stress response                    3.488015e-21\n",
       "Apoptosis                          3.809927e-23\n",
       "PCP pathway                        3.809927e-23\n",
       "Lipid homeostasis                  3.809927e-23\n",
       "Immune response                    1.686896e-16\n",
       "Epidermal development              3.809927e-23\n",
       "Longevity                          5.294398e-06\n",
       "dtype: float64"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wilcoxon_functions.loc[sorted_order, ['Greater Q-val', 'Less Q-val']].min(axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Apoptosis\n",
      "Canonical RTK-Ras-ERK signaling\n",
      "Cell adhesion\n",
      "Cell migration\n",
      "Eph receptor signaling\n",
      "Epidermal development\n",
      "Hedgehog signaling\n",
      "Immune response\n",
      "Insulin signaling\n",
      "Lipid homeostasis\n",
      "Longevity\n",
      "Mechanosensory mechanism\n",
      "Neuronal communication\n",
      "Notch signaling\n",
      "PCP pathway\n",
      "Receptor-Mediated-Endocytosis\n",
      "Stress response\n",
      "TGF-B signaling\n",
      "Trypsin-like proteolysis\n",
      "Wnt signaling\n"
     ]
    }
   ],
   "source": [
    "print(*wilcoxon_functions.index, sep='\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}