kg.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys, os, re\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import json\n",
    "import requests\n",
    "import csv\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "string_name = pd.read_csv('data/9606.protein.info.v11.5.txt', sep=\"\\t\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>#string_protein_id</th>\n",
       "      <th>preferred_name</th>\n",
       "      <th>protein_size</th>\n",
       "      <th>annotation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9606.ENSP00000000233</td>\n",
       "      <td>ARF5</td>\n",
       "      <td>180</td>\n",
       "      <td>ADP-ribosylation factor 5; GTP-binding protein...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9606.ENSP00000000412</td>\n",
       "      <td>M6PR</td>\n",
       "      <td>277</td>\n",
       "      <td>Cation-dependent mannose-6-phosphate receptor;...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9606.ENSP00000001008</td>\n",
       "      <td>FKBP4</td>\n",
       "      <td>459</td>\n",
       "      <td>Peptidyl-prolyl cis-trans isomerase FKBP4; Imm...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9606.ENSP00000001146</td>\n",
       "      <td>CYP26B1</td>\n",
       "      <td>512</td>\n",
       "      <td>Cytochrome P450 26B1; Involved in the metaboli...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>9606.ENSP00000002125</td>\n",
       "      <td>NDUFAF7</td>\n",
       "      <td>441</td>\n",
       "      <td>Protein arginine methyltransferase NDUFAF7, mi...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     #string_protein_id preferred_name  protein_size  \\\n",
       "0  9606.ENSP00000000233           ARF5           180   \n",
       "1  9606.ENSP00000000412           M6PR           277   \n",
       "2  9606.ENSP00000001008          FKBP4           459   \n",
       "3  9606.ENSP00000001146        CYP26B1           512   \n",
       "4  9606.ENSP00000002125        NDUFAF7           441   \n",
       "\n",
       "                                          annotation  \n",
       "0  ADP-ribosylation factor 5; GTP-binding protein...  \n",
       "1  Cation-dependent mannose-6-phosphate receptor;...  \n",
       "2  Peptidyl-prolyl cis-trans isomerase FKBP4; Imm...  \n",
       "3  Cytochrome P450 26B1; Involved in the metaboli...  \n",
       "4  Protein arginine methyltransferase NDUFAF7, mi...  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "string_name.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "STRING_df_dict = {}\n",
    "\n",
    "string_name_dict = string_name.to_dict('split')\n",
    "\n",
    "for i in range(len(string_name_dict['data'])):\n",
    "    STRING_df_dict[string_name_dict['data'][i][0]] = {\"name\": string_name_dict['data'][i][1], \"description\": string_name_dict['data'][i][3].split(\";\")[0].strip()}\n",
    "\n",
    "del string_name, string_name_dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>protein1</th>\n",
       "      <th>protein2</th>\n",
       "      <th>neighborhood</th>\n",
       "      <th>fusion</th>\n",
       "      <th>cooccurence</th>\n",
       "      <th>coexpression</th>\n",
       "      <th>experimental</th>\n",
       "      <th>database</th>\n",
       "      <th>textmining</th>\n",
       "      <th>combined_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9606.ENSP00000000233</td>\n",
       "      <td>9606.ENSP00000379496</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>54</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>144</td>\n",
       "      <td>155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9606.ENSP00000000233</td>\n",
       "      <td>9606.ENSP00000314067</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>180</td>\n",
       "      <td>0</td>\n",
       "      <td>61</td>\n",
       "      <td>197</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9606.ENSP00000000233</td>\n",
       "      <td>9606.ENSP00000263116</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>62</td>\n",
       "      <td>152</td>\n",
       "      <td>0</td>\n",
       "      <td>101</td>\n",
       "      <td>222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9606.ENSP00000000233</td>\n",
       "      <td>9606.ENSP00000361263</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>161</td>\n",
       "      <td>0</td>\n",
       "      <td>64</td>\n",
       "      <td>181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>9606.ENSP00000000233</td>\n",
       "      <td>9606.ENSP00000409666</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>82</td>\n",
       "      <td>213</td>\n",
       "      <td>0</td>\n",
       "      <td>72</td>\n",
       "      <td>270</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               protein1              protein2  neighborhood  fusion  \\\n",
       "0  9606.ENSP00000000233  9606.ENSP00000379496             0       0   \n",
       "1  9606.ENSP00000000233  9606.ENSP00000314067             0       0   \n",
       "2  9606.ENSP00000000233  9606.ENSP00000263116             0       0   \n",
       "3  9606.ENSP00000000233  9606.ENSP00000361263             0       0   \n",
       "4  9606.ENSP00000000233  9606.ENSP00000409666             0       0   \n",
       "\n",
       "   cooccurence  coexpression  experimental  database  textmining  \\\n",
       "0            0            54             0         0         144   \n",
       "1            0             0           180         0          61   \n",
       "2            0            62           152         0         101   \n",
       "3            0             0           161         0          64   \n",
       "4            0            82           213         0          72   \n",
       "\n",
       "   combined_score  \n",
       "0             155  \n",
       "1             197  \n",
       "2             222  \n",
       "3             181  \n",
       "4             270  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "##string\n",
    "\n",
    "string_link = pd.read_csv('data/9606.protein.links.detailed.v11.5.txt', sep=\" \")\n",
    "\n",
    "string_link.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "combine_cutoff_score = 900\n",
    "\n",
    "string_link_dict = string_link[string_link[\"combined_score\"] >= combine_cutoff_score].to_dict('split')\n",
    "\n",
    "link_set = set()\n",
    "\n",
    "STRING_to_add = {}\n",
    "\n",
    "for i in range(len(string_link_dict['data'])):\n",
    "    protein_1 = string_link_dict['data'][i][0]\n",
    "    protein_2 = string_link_dict['data'][i][1]\n",
    "\n",
    "    #print (protein_1, protein_2)\n",
    "\n",
    "    if \"name\" in STRING_df_dict[protein_1] and \"name\" in STRING_df_dict[protein_2]:\n",
    "        protein_name_1 = STRING_df_dict[protein_1][\"name\"]\n",
    "        protein_name_2 = STRING_df_dict[protein_2][\"name\"]\n",
    "\n",
    "        t = tuple(sorted([protein_name_1, protein_name_2]))\n",
    "        link_set.add(t)\n",
    "\n",
    "        STRING_to_add[protein_name_1.lower()] = protein_name_1\n",
    "        STRING_to_add[protein_name_2.lower()] = protein_name_2\n",
    "    else:\n",
    "        print (\"No name for protein: \", protein_1, protein_2)\n",
    "del string_link_dict, string_link"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "123577"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(link_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "11813"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(STRING_to_add)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now correct the names in hetionet.gene to STRING"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "json_connections = set()\n",
    "\n",
    "\n",
    "with open('data/hetionet-v1.0.json', 'r') as f:\n",
    "    json_data = json.load(f)\n",
    "\n",
    "    id_name = {}\n",
    "\n",
    "    nodes = []\n",
    "    for g in json_data[\"nodes\"]:\n",
    "        if g[\"kind\"] == \"Gene\":\n",
    "            if g[\"name\"].lower() in STRING_to_add:\n",
    "                g[\"name\"] = STRING_to_add[g[\"name\"].lower()]\n",
    "\n",
    "        id_name[g[\"identifier\"]] = g[\"name\"]\n",
    "        \n",
    "        if \"data\" in g:\n",
    "            del g[\"data\"]\n",
    "        nodes.append(g)\n",
    "\n",
    "    with open('data/hetionet_string_corrected_gene_name_nodes.json', 'w') as outfile:\n",
    "        json.dump(nodes, outfile, indent=2)\n",
    "\n",
    "    edges = []\n",
    "    for edge in json_data[\"edges\"]:\n",
    "        id_1 = edge[\"source_id\"][1]\n",
    "        name_1 = id_name[id_1]\n",
    "        edge[\"source_id\"][1] = name_1\n",
    "        \n",
    "        id_2 = edge[\"target_id\"][1]\n",
    "        name_2 = id_name[id_2]\n",
    "        edge[\"target_id\"][1] = name_2\n",
    "\n",
    "        if \"data\" in edge:\n",
    "            del edge[\"data\"]\n",
    "    \n",
    "        edges.append(edge)\n",
    "\n",
    "    with open('data/hetionet_string_corrected_gene_name_edges.json', 'w') as outfile:\n",
    "        json.dump(edges, outfile, indent=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now correct the names with kegg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"data/disease.csv\")\n",
    "\n",
    "string_name_dict = df.to_dict('split')\n",
    "\n",
    "kegg_disease_name = {}\n",
    "\n",
    "for i in range(len(string_name_dict['data'])):\n",
    "    kegg_disease_name[string_name_dict['data'][i][1].lower()] = string_name_dict['data'][i][1]\n",
    "\n",
    "\n",
    "df = pd.read_csv(\"data/drug.csv\")\n",
    "\n",
    "string_name_dict = df.to_dict('split')\n",
    "\n",
    "kegg_compound_name = {}\n",
    "\n",
    "for i in range(len(string_name_dict['data'])):\n",
    "    kegg_compound_name[string_name_dict['data'][i][1].lower()] = string_name_dict['data'][i][1]\n",
    "\n",
    "\n",
    "df = pd.read_csv(\"data/gene_tmp.csv\")\n",
    "\n",
    "string_name_dict = df.to_dict('split')\n",
    "\n",
    "kegg_gene_name = {}\n",
    "\n",
    "for i in range(len(string_name_dict['data'])):\n",
    "    kegg_gene_name[string_name_dict['data'][i][1].lower()] = string_name_dict['data'][i][1]\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "with open('data/hetionet_string_corrected_gene_name_nodes.json', 'r') as f:\n",
    "    nodes = json.load(f)\n",
    "\n",
    "    new_nodes = []\n",
    "    for node in nodes:\n",
    "        if node[\"kind\"] == \"Disease\":\n",
    "            if node[\"name\"].lower() in kegg_disease_name:\n",
    "                node[\"name\"] = kegg_disease_name[node[\"name\"].lower()]\n",
    "        elif node[\"kind\"] == \"Compound\":\n",
    "            if node[\"name\"].lower() in kegg_compound_name:\n",
    "                node[\"name\"] = kegg_compound_name[node[\"name\"].lower()]\n",
    "        elif node[\"kind\"] == \"Gene\":\n",
    "            if node[\"name\"].lower() in kegg_gene_name:\n",
    "                node[\"name\"] = kegg_gene_name[node[\"name\"].lower()]\n",
    "        \n",
    "        new_nodes.append(node)\n",
    "\n",
    "        \n",
    "    with open('data/hetionet_kegg_corrected_gene_name_nodes.json', 'w') as outfile:\n",
    "        json.dump(new_nodes, outfile, indent=2)\n",
    "\n",
    "\n",
    "with open('data/hetionet_string_corrected_gene_name_edges.json', 'r') as f:\n",
    "    edges = json.load(f)\n",
    "\n",
    "    new_edges = []\n",
    "    for edge in edges:\n",
    "        if edge[\"source_id\"][0] == \"Disease\":\n",
    "            if edge[\"source_id\"][1].lower() in kegg_disease_name:\n",
    "                edge[\"source_id\"][1] = kegg_disease_name[edge[\"source_id\"][1].lower()]\n",
    "\n",
    "        if edge[\"target_id\"][0] == \"Disease\":\n",
    "            if edge[\"target_id\"][1].lower() in kegg_disease_name:\n",
    "                edge[\"target_id\"][1] = kegg_disease_name[edge[\"target_id\"][1].lower()]\n",
    "        \n",
    "\n",
    "        if edge[\"source_id\"][0] == \"Compound\":\n",
    "            if edge[\"source_id\"][1].lower() in kegg_compound_name:\n",
    "                edge[\"source_id\"][1] = kegg_compound_name[edge[\"source_id\"][1].lower()]\n",
    "\n",
    "        if edge[\"target_id\"][0] == \"Compound\":\n",
    "            if edge[\"target_id\"][1].lower() in kegg_compound_name:\n",
    "                edge[\"target_id\"][1] = kegg_compound_name[edge[\"target_id\"][1].lower()]\n",
    "\n",
    "        \n",
    "        if edge[\"source_id\"][0] == \"Gene\":\n",
    "            if edge[\"source_id\"][1].lower() in kegg_gene_name:\n",
    "                edge[\"source_id\"][1] = kegg_gene_name[edge[\"source_id\"][1].lower()]\n",
    "\n",
    "        if edge[\"target_id\"][0] == \"Gene\":\n",
    "            if edge[\"target_id\"][1].lower() in kegg_gene_name:\n",
    "                edge[\"target_id\"][1] = kegg_gene_name[edge[\"target_id\"][1].lower()]\n",
    "        \n",
    "        new_edges.append(edge)\n",
    "\n",
    "        \n",
    "    with open('data/hetionet_kegg_corrected_gene_name_edges.json', 'w') as outfile:\n",
    "        json.dump(new_edges, outfile, indent=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "save string links and nodes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19566\n",
      "      name                                        description\n",
      "0     ARF5                          ADP-ribosylation factor 5\n",
      "1     M6PR      Cation-dependent mannose-6-phosphate receptor\n",
      "2    FKBP4          Peptidyl-prolyl cis-trans isomerase FKBP4\n",
      "3  CYP26B1                               Cytochrome P450 26B1\n",
      "4  NDUFAF7  Protein arginine methyltransferase NDUFAF7, mi...\n",
      "11814\n"
     ]
    }
   ],
   "source": [
    "STRING_to_add.values()\n",
    "string_name = pd.read_csv('data/9606.protein.info.v11.5.txt', sep=\"\\t\")\n",
    "\n",
    "print (len(string_name))\n",
    "#df_kegg_disease_diff_hetionet = df_kegg_disease[~df_kegg_disease[\"name\"].str.lower().isin(hetionet_disease_name)]\n",
    "#print (string_name.head())\n",
    "\n",
    "string_name = string_name[string_name[\"preferred_name\"].isin(STRING_to_add.values())]\n",
    "string_name[\"description\"] = string_name[\"annotation\"].str.split(\";\").str[0]\n",
    "\n",
    "string_name.drop([\"#string_protein_id\", \"protein_size\", \"annotation\"], axis = 1, inplace = True)\n",
    "\n",
    "string_name.rename({'preferred_name': 'name'}, axis='columns', inplace=True)\n",
    "\n",
    "print (string_name.head())\n",
    "\n",
    "print (len(string_name))\n",
    "\n",
    "string_name.to_csv(\"data/string_to_add_nodes.csv\", index = False, quoting=csv.QUOTE_ALL)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "content = []\n",
    "\"\"\"\n",
    "{\n",
    "    \"source_id\": [\n",
    "      \"Gene\",\n",
    "      \"CALM1\"\n",
    "    ],\n",
    "    \"target_id\": [\n",
    "      \"Gene\",\n",
    "      \"VHL\"\n",
    "    ],\n",
    "    \"kind\": \"interacts\",\n",
    "    \"direction\": \"both\"\n",
    "  },\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "for new_edge in link_set:\n",
    "    content.append({\"source_id\": [\"Gene\", new_edge[0]], \"target_id\": [\"Gene\", new_edge[1]], \"kind\": \"STRING\"})\n",
    "\n",
    "with open('data/string_to_add_edges.json', 'w') as outfile:\n",
    "    json.dump(content, outfile, indent=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "970f8b811fa5a8347730b7014810b8a3f431f32c1d447a94b9c225bb540cbb7a"
  },
  "kernelspec": {
   "display_name": "Python 3.8.8 64-bit ('base': conda)",
   "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": "3.8.5"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}