bind_expr_filters.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Set B.1.351 RBD DMS ACE2 binding and expression scores for serum mapping\n",
    "We want to make sure that the filters chosen for the ACE2 binding and RBD expression scores are reasonable such that spurious antibody-escpae mutations that merely fall into the antibody-escape gate due to their poor folding or expression are removed. \n",
    "\n",
    "But, we also want to make sure we aren't throwing out many mutations that are found in nature at reasonable numbers. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "from IPython.display import display, HTML\n",
    "\n",
    "import math\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from scipy import stats\n",
    "\n",
    "from plotnine import *\n",
    "\n",
    "from dms_variants.constants import CBPALETTE\n",
    "\n",
    "import yaml"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Read config file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('config.yaml') as f:\n",
    "    config = yaml.safe_load(f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Define input and output directories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "datadir = 'data'\n",
    "resultsdir = config['bind_expr_filters_dir']\n",
    "\n",
    "os.makedirs(resultsdir, exist_ok=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Read in the new filters for DMS ACE2 binding and expression scores. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>metric</th>\n",
       "      <th>score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>delta_bind</td>\n",
       "      <td>-3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>delta_expr</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "og_thresholds={'delta_bind':-2.35, 'delta_expr':-1.0}\n",
    "new_thresholds={'delta_bind':config['escape_score_min_bind_mut'], 'delta_expr':config['escape_score_min_expr_mut']}\n",
    "\n",
    "og_thresholds_df=pd.DataFrame.from_dict({'metric': ['delta_bind', 'delta_expr'], 'score': [-2.35,-1.0]})\n",
    "new_filter_df=pd.DataFrame({'metric': ['delta_bind', 'delta_expr'], 'score':[config['escape_score_min_bind_mut'],config['escape_score_min_expr_mut']]})\n",
    "display(HTML(new_filter_df.to_html(index=False)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "gisaid_counts_file = config['gisaid_mutation_counts']\n",
    "dms_scores_file = config['prelim_mut_bind_expr']\n",
    "og_dms_file = config['early2020_mut_bind_expr']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### First, just plot the correlations between libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "15276\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>wildtype</th>\n",
       "      <th>mutation</th>\n",
       "      <th>site</th>\n",
       "      <th>mutant</th>\n",
       "      <th>score</th>\n",
       "      <th>metric</th>\n",
       "      <th>library</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331A</td>\n",
       "      <td>331</td>\n",
       "      <td>A</td>\n",
       "      <td>0.10425</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>lib1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331C</td>\n",
       "      <td>331</td>\n",
       "      <td>C</td>\n",
       "      <td>-0.14457</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>lib1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331D</td>\n",
       "      <td>331</td>\n",
       "      <td>D</td>\n",
       "      <td>-0.02461</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>lib1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331E</td>\n",
       "      <td>331</td>\n",
       "      <td>E</td>\n",
       "      <td>0.14659</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>lib1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331F</td>\n",
       "      <td>331</td>\n",
       "      <td>F</td>\n",
       "      <td>-0.13788</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>lib1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dms_corrs = pd.DataFrame()\n",
    "\n",
    "for lib in ['lib1', 'lib2']:\n",
    "    for m in ['bind', 'expr']:\n",
    "        \n",
    "        df = (pd.read_csv(dms_scores_file).rename(columns={'position': 'site'})\n",
    "                     [['wildtype', 'mutation', 'site', 'mutant', f'delta_{m}_{lib}']]\n",
    "                     .assign(not_WT=lambda x: x['wildtype']!=x['mutant'])\n",
    "                     .query('not_WT')\n",
    "                     .drop(columns=['not_WT'])\n",
    "                     .assign(metric=f'delta_{m}',\n",
    "                             library=lib,\n",
    "                            )\n",
    "              .rename(columns={f'delta_{m}_{lib}': 'score'})\n",
    "                     )\n",
    "        dms_corrs = pd.concat([dms_corrs, df])\n",
    "\n",
    "print(len(dms_corrs))\n",
    "display(HTML(dms_corrs.head().to_html(index=False)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_8465/218136499.py:15: FutureWarning: Indexing with multiple keys (implicitly converted to a tuple of keys) will be deprecated, use a list instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving plot to results/bind_expr_filters/replicates-corr.pdf\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x250 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# wide data frame with each library's score in a different column\n",
    "df_wide = (\n",
    "    dms_corrs\n",
    "    .replace({'delta_bind': 'ACE2 binding', 'delta_expr': 'RBD expression'})\n",
    "    # now make columns for each library, only keep mutants with scores for both libs\n",
    "    .pivot_table(index=['mutation', 'metric'],\n",
    "                 columns='library',\n",
    "                 values='score')\n",
    "    .reset_index()\n",
    "    .dropna(axis=0)\n",
    "    )\n",
    "\n",
    "# correlations between libraries\n",
    "corrs = (\n",
    "    df_wide\n",
    "    .groupby('metric')\n",
    "    ['lib1', 'lib2']\n",
    "    .corr(method='pearson')\n",
    "    .reset_index()\n",
    "    .query('library == \"lib1\"')\n",
    "    .assign(correlation=lambda x: 'R=' + x['lib2'].round(2).astype(str))\n",
    "    [['metric', 'correlation']]\n",
    "    # add number of mutations measured\n",
    "    .merge(df_wide\n",
    "           .groupby('metric')\n",
    "           .size()\n",
    "           .rename('n')\n",
    "           .reset_index()\n",
    "           )\n",
    "    .assign(correlation=lambda x: x['correlation'] + ', N=' + x['n'].astype(str))\n",
    "    )\n",
    "    \n",
    "# plot correlations\n",
    "xmin = df_wide['lib1'].min()\n",
    "xspan = df_wide['lib1'].max() - xmin\n",
    "ymin = df_wide['lib2'].min()\n",
    "yspan = df_wide['lib2'].max() - ymin\n",
    "\n",
    "p = (ggplot(df_wide) +\n",
    "     aes('lib1', 'lib2') +\n",
    "     geom_point(alpha=0.2, color='black') +\n",
    "     geom_text(mapping=aes(label='correlation'),\n",
    "               data=corrs,\n",
    "               x=0.01 * xspan + xmin,\n",
    "               y=0.99 * yspan + ymin,\n",
    "               size=10,\n",
    "               ha='left',\n",
    "               va='top',\n",
    "               ) +\n",
    "     facet_wrap('~ metric') +\n",
    "     theme_classic() +\n",
    "     theme(figure_size=(2.5 * 2, 2.5 * 1),\n",
    "           plot_title=element_text(size=14),\n",
    "           strip_background=element_blank(),\n",
    "           strip_text=element_text(size=12),\n",
    "          )\n",
    "     )\n",
    "\n",
    "p.draw()\n",
    "\n",
    "plotfile = os.path.join(resultsdir, f\"replicates-corr.pdf\")\n",
    "print(f\"Saving plot to {plotfile}\")\n",
    "p.save(plotfile, verbose=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Examine filters and GISAID counts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>wildtype</th>\n",
       "      <th>mutation</th>\n",
       "      <th>site</th>\n",
       "      <th>mutant</th>\n",
       "      <th>delta_bind</th>\n",
       "      <th>delta_expr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331A</td>\n",
       "      <td>331</td>\n",
       "      <td>A</td>\n",
       "      <td>0.12426</td>\n",
       "      <td>-0.27844</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331C</td>\n",
       "      <td>331</td>\n",
       "      <td>C</td>\n",
       "      <td>-0.20203</td>\n",
       "      <td>-0.63606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331D</td>\n",
       "      <td>331</td>\n",
       "      <td>D</td>\n",
       "      <td>0.00131</td>\n",
       "      <td>-0.22296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331E</td>\n",
       "      <td>331</td>\n",
       "      <td>E</td>\n",
       "      <td>0.10980</td>\n",
       "      <td>0.06430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331F</td>\n",
       "      <td>331</td>\n",
       "      <td>F</td>\n",
       "      <td>-0.15674</td>\n",
       "      <td>-0.57560</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dms_scores = (pd.read_csv(dms_scores_file).rename(columns={'position': 'site'})\n",
    "             [['wildtype', 'mutation', 'site', 'mutant', 'delta_bind', 'delta_expr']]\n",
    "             )\n",
    "\n",
    "display(HTML(dms_scores.head().to_html(index=False)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>wildtype</th>\n",
       "      <th>mutation</th>\n",
       "      <th>site</th>\n",
       "      <th>mutant</th>\n",
       "      <th>count</th>\n",
       "      <th>n_countries</th>\n",
       "      <th>frequency</th>\n",
       "      <th>metric</th>\n",
       "      <th>score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331A</td>\n",
       "      <td>331</td>\n",
       "      <td>A</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>0.12426</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331C</td>\n",
       "      <td>331</td>\n",
       "      <td>C</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>-0.20203</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331D</td>\n",
       "      <td>331</td>\n",
       "      <td>D</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>8.673362e-07</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>0.00131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331E</td>\n",
       "      <td>331</td>\n",
       "      <td>E</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>0.10980</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331F</td>\n",
       "      <td>331</td>\n",
       "      <td>F</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>-0.15674</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "gisaid_counts = (pd.read_csv(gisaid_counts_file)\n",
    "                 .drop(columns=['isite', 'wildtype'])\n",
    "                )\n",
    "\n",
    "dms_scores=(dms_scores\n",
    "            .merge(gisaid_counts,\n",
    "                   on=['site', 'mutant'],\n",
    "                   how='left',\n",
    "                   validate='many_to_one',\n",
    "                  )\n",
    "            .fillna({'count':0,'n_countries':0, 'frequency': 0})\n",
    "           )\n",
    "\n",
    "dms_scores=dms_scores.melt(id_vars=['wildtype','mutation', 'site', 'mutant', 'count', 'n_countries', 'frequency'],\n",
    "                           value_vars=['delta_bind', 'delta_expr'], \n",
    "                           var_name='metric', \n",
    "                           value_name='score',\n",
    "                          )\n",
    "\n",
    "display(HTML(dms_scores.head().to_html(index=False)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving plot to results/bind_expr_filters/counts-v-score.pdf\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x250 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p = (ggplot(dms_scores\n",
    "            # assign small numbers to things with 0 GISAID counts or missing scores so they still appear on plot \n",
    "            .replace({'count': {0: 0.1}, 'score': {np.nan: -5}})\n",
    "            .replace({'delta_bind':'ACE2 binding', 'delta_expr':'RBD expression'})\n",
    "           ) +\n",
    "     aes('count', 'score') +\n",
    "     geom_point(alpha=0.2, color='black') +\n",
    "     facet_grid('~ metric') +\n",
    "     scale_x_log10()+\n",
    "     theme_classic() +\n",
    "     geom_hline(data=new_filter_df.replace({'delta_bind':'ACE2 binding', 'delta_expr':'RBD expression'}),\n",
    "                 mapping=aes(yintercept='score'),\n",
    "                linetype='dashed',\n",
    "                color=CBPALETTE[1])+\n",
    "     theme(figure_size=(2.5 * 2, 2.5 * 1),\n",
    "           strip_background=element_blank(),\n",
    "           strip_text=element_text(size=12),\n",
    "          ) +\n",
    "     xlab('mutation counts in GISAID as of Aug. 1, 2021')+\n",
    "     ylab('B.1.351 RBD DMS score\\n(single mutants)')\n",
    "     )\n",
    "\n",
    "fig = p.draw()\n",
    "\n",
    "plotfile = os.path.join(resultsdir, f\"counts-v-score.pdf\")\n",
    "print(f\"Saving plot to {plotfile}\")\n",
    "p.save(plotfile, verbose=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/fh/fast/bloom_j/computational_notebooks/agreaney/2021/SARS-CoV-2-RBD_B.1.351/env/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_boxplot : Removed 34 rows containing non-finite values.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving plot to results/bind_expr_filters/count-cat-v-score.pdf\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/fh/fast/bloom_j/computational_notebooks/agreaney/2021/SARS-CoV-2-RBD_B.1.351/env/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_boxplot : Removed 34 rows containing non-finite values.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x250 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def assign_count_categories(x):\n",
    "    if x == 0:\n",
    "        return \"0\"\n",
    "    elif x < 10:\n",
    "        return \"1 to 9\"\n",
    "    elif x < 20:\n",
    "        return \"10 to 19\"\n",
    "    elif x < 50:\n",
    "        return \"20 to 49\"\n",
    "    else:\n",
    "        return \">=50\"\n",
    "    \n",
    "count_categories=[\"0\", \"1 to 9\", \"10 to 19\", \"20 to 49\", \">=50\"]\n",
    "\n",
    "dms_scores=(dms_scores\n",
    "            .assign(count_categories=lambda x: x['count'].apply(assign_count_categories),\n",
    "                   )\n",
    "           )\n",
    "\n",
    "dms_scores=(dms_scores\n",
    "            .assign(count_categories=lambda x: pd.Categorical(x['count_categories'],\n",
    "                                                              categories=count_categories,\n",
    "                                                              ordered=True\n",
    "                                                             ))\n",
    "           )\n",
    "\n",
    "p = (ggplot(dms_scores.replace({'delta_bind':'ACE2 binding', 'delta_expr':'RBD expression'})) +\n",
    "     aes('count_categories', 'score') +\n",
    "     geom_hline(data=new_filter_df.replace({'delta_bind':'ACE2 binding', 'delta_expr':'RBD expression'}),\n",
    "                 mapping=aes(yintercept='score'),\n",
    "                linetype='dashed',\n",
    "                color=CBPALETTE[1])+\n",
    "     geom_boxplot(outlier_alpha=0.2) +\n",
    "     facet_grid('~ metric') +\n",
    "     theme_classic() +\n",
    "     theme(figure_size=(2.5 * 2, 2.5 * 1),\n",
    "           axis_text_x=element_text(angle=90),\n",
    "           strip_background=element_blank(),\n",
    "           strip_text=element_text(size=12),\n",
    "          ) +\n",
    "     xlab('mutation counts in GISAID as of Aug. 1 2021')+\n",
    "     ylab('B.1.351 RBD DMS score')\n",
    "     )\n",
    "\n",
    "fig = p.draw()\n",
    "\n",
    "plotfile = os.path.join(resultsdir, f\"count-cat-v-score.pdf\")\n",
    "print(f\"Saving plot to {plotfile}\")\n",
    "p.save(plotfile, verbose=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/fh/fast/bloom_j/computational_notebooks/agreaney/2021/SARS-CoV-2-RBD_B.1.351/env/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_bin : Removed 55 rows containing non-finite values.\n",
      "/fh/fast/bloom_j/computational_notebooks/agreaney/2021/SARS-CoV-2-RBD_B.1.351/env/lib/python3.8/site-packages/plotnine/layer.py:401: PlotnineWarning: geom_histogram : Removed 20 rows containing missing values.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving plot to results/bind_expr_filters/count-score-histogram.pdf\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/fh/fast/bloom_j/computational_notebooks/agreaney/2021/SARS-CoV-2-RBD_B.1.351/env/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_bin : Removed 55 rows containing non-finite values.\n",
      "/fh/fast/bloom_j/computational_notebooks/agreaney/2021/SARS-CoV-2-RBD_B.1.351/env/lib/python3.8/site-packages/plotnine/layer.py:401: PlotnineWarning: geom_histogram : Removed 20 rows containing missing values.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x250 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "x_min=-4.5\n",
    "x_max=0.5\n",
    "\n",
    "p = (ggplot(dms_scores.replace({'delta_bind':'ACE2 binding', 'delta_expr':'RBD expression'})) +\n",
    "     aes(x='score', fill='count_categories') +\n",
    "     geom_histogram(position='identity', bins=50) +\n",
    "     facet_grid('~ metric') +\n",
    "     scale_x_continuous(breaks=np.arange(x_min,x_max,0.5), limits=[x_min, x_max]) +\n",
    "     geom_vline(data=new_filter_df.replace({'delta_bind':'ACE2 binding', 'delta_expr':'RBD expression'}),\n",
    "                     mapping=aes(xintercept='score'),\n",
    "                    linetype='dashed',\n",
    "                    color=CBPALETTE[1])+\n",
    "     theme_classic() +\n",
    "     theme(figure_size=(2.5 * 2, 2.5 * 1),\n",
    "           plot_title=element_text(size=14),\n",
    "           axis_text_x=element_text(angle=90),\n",
    "           strip_background=element_blank(),\n",
    "           strip_text=element_text(size=12),\n",
    "          ) +\n",
    "     ylab('number of mutations')+\n",
    "     xlab('B.1.351 RBD DMS score') +\n",
    "     labs(fill='GISAID counts')\n",
    "     )\n",
    "\n",
    "fig = p.draw()\n",
    "\n",
    "plotfile = os.path.join(resultsdir, f\"count-score-histogram.pdf\")\n",
    "print(f\"Saving plot to {plotfile}\")\n",
    "p.save(plotfile, verbose=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/fh/fast/bloom_j/computational_notebooks/agreaney/2021/SARS-CoV-2-RBD_B.1.351/env/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_density : Removed 21 rows containing non-finite values.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving plot to results/bind_expr_filters/count-score-density.pdf\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/fh/fast/bloom_j/computational_notebooks/agreaney/2021/SARS-CoV-2-RBD_B.1.351/env/lib/python3.8/site-packages/plotnine/layer.py:324: PlotnineWarning: stat_density : Removed 21 rows containing non-finite values.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x250 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p = (ggplot(dms_scores\n",
    "            .replace({'delta_bind':'ACE2 binding', 'delta_expr':'RBD expression'})\n",
    "            .dropna()\n",
    "            .drop(columns=['count'])\n",
    "            .assign(count_categories=lambda x: pd.Categorical(x['count_categories']).remove_unused_categories())\n",
    "           ) +\n",
    "     aes(x='score', fill='count_categories') +\n",
    "     geom_density(alpha=0.5) +#aes(y=after_stat('count')), \n",
    "     scale_x_continuous(breaks=np.arange(x_min,x_max,0.5), limits=[x_min, x_max]) +\n",
    "     geom_vline(data=new_filter_df.replace({'delta_bind':'ACE2 binding', 'delta_expr':'RBD expression'}),\n",
    "                     mapping=aes(xintercept='score'),\n",
    "                    linetype='dashed',\n",
    "                    color=CBPALETTE[1])+\n",
    "     facet_grid('~ metric') +\n",
    "     theme_classic() +\n",
    "     theme(figure_size=(2.5 * 2, 2.5 * 1),\n",
    "           plot_title=element_text(size=14),\n",
    "           axis_text_x=element_text(angle=90),\n",
    "           strip_background=element_blank(),\n",
    "           strip_text=element_text(size=12),\n",
    "          ) +\n",
    "     labs(fill = 'GISAID counts')+\n",
    "     xlab('B.1.351 RBD DMS score')\n",
    "     )\n",
    "\n",
    "fig = p.draw()\n",
    "\n",
    "plotfile = os.path.join(resultsdir, f\"count-score-density.pdf\")\n",
    "print(f\"Saving plot to {plotfile}\")\n",
    "p.save(plotfile, verbose=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Things I want to know:\n",
    "1. Mutations that have **any** counts in nature but are missing scores\n",
    "2. Mutations that have appreciable counts (>=50) in nature but very low scores\n",
    "3. The scores corresponding to the 95th percentile of all mutations occurring >= 50x in nature\n",
    "4. The scores of mutations to disulfide bonds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Here are the naturally occurring mutations that are missing scores from B.1.351 DMS\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>wildtype</th>\n",
       "      <th>mutation</th>\n",
       "      <th>count</th>\n",
       "      <th>n_countries</th>\n",
       "      <th>frequency</th>\n",
       "      <th>score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>V</td>\n",
       "      <td>V341F</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.301004e-06</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>V</td>\n",
       "      <td>V483L</td>\n",
       "      <td>33.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>1.431105e-05</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>K</td>\n",
       "      <td>K529I</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.336681e-07</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>G</td>\n",
       "      <td>G526A</td>\n",
       "      <td>11.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.770349e-06</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Here are the naturally occurring mutations that are missing scores from B.1.351 DMS')\n",
    "display(HTML(dms_scores\n",
    "             .query('count >= 1')\n",
    "             .query('score.isnull()', engine='python')\n",
    "             [['wildtype','mutation', 'count', 'n_countries', 'frequency', 'score']]\n",
    "             .drop_duplicates()\n",
    "             .to_html(index=False)\n",
    "            )\n",
    "       )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mutations with >=50 GISAID counts but with bind score < -3.0\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>wildtype</th>\n",
       "      <th>mutation</th>\n",
       "      <th>site</th>\n",
       "      <th>mutant</th>\n",
       "      <th>count</th>\n",
       "      <th>n_countries</th>\n",
       "      <th>frequency</th>\n",
       "      <th>metric</th>\n",
       "      <th>score</th>\n",
       "      <th>count_categories</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>A</td>\n",
       "      <td>A419S</td>\n",
       "      <td>419</td>\n",
       "      <td>S</td>\n",
       "      <td>134.0</td>\n",
       "      <td>23.0</td>\n",
       "      <td>0.000058</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>-4.07656</td>\n",
       "      <td>&gt;=50</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mutations with >=50 GISAID counts but with expr score < -1.0\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>wildtype</th>\n",
       "      <th>mutation</th>\n",
       "      <th>site</th>\n",
       "      <th>mutant</th>\n",
       "      <th>count</th>\n",
       "      <th>n_countries</th>\n",
       "      <th>frequency</th>\n",
       "      <th>metric</th>\n",
       "      <th>score</th>\n",
       "      <th>count_categories</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>T</td>\n",
       "      <td>T470N</td>\n",
       "      <td>470</td>\n",
       "      <td>N</td>\n",
       "      <td>189.0</td>\n",
       "      <td>23.0</td>\n",
       "      <td>0.000082</td>\n",
       "      <td>delta_expr</td>\n",
       "      <td>-2.68080</td>\n",
       "      <td>&gt;=50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>V</td>\n",
       "      <td>V401L</td>\n",
       "      <td>401</td>\n",
       "      <td>L</td>\n",
       "      <td>107.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>0.000046</td>\n",
       "      <td>delta_expr</td>\n",
       "      <td>-2.01452</td>\n",
       "      <td>&gt;=50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>G</td>\n",
       "      <td>G404C</td>\n",
       "      <td>404</td>\n",
       "      <td>C</td>\n",
       "      <td>55.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.000024</td>\n",
       "      <td>delta_expr</td>\n",
       "      <td>-1.98436</td>\n",
       "      <td>&gt;=50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>A</td>\n",
       "      <td>A419S</td>\n",
       "      <td>419</td>\n",
       "      <td>S</td>\n",
       "      <td>134.0</td>\n",
       "      <td>23.0</td>\n",
       "      <td>0.000058</td>\n",
       "      <td>delta_expr</td>\n",
       "      <td>-1.77949</td>\n",
       "      <td>&gt;=50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>A</td>\n",
       "      <td>A352V</td>\n",
       "      <td>352</td>\n",
       "      <td>V</td>\n",
       "      <td>79.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>0.000034</td>\n",
       "      <td>delta_expr</td>\n",
       "      <td>-1.62895</td>\n",
       "      <td>&gt;=50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>G</td>\n",
       "      <td>G485V</td>\n",
       "      <td>485</td>\n",
       "      <td>V</td>\n",
       "      <td>60.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>0.000026</td>\n",
       "      <td>delta_expr</td>\n",
       "      <td>-1.24302</td>\n",
       "      <td>&gt;=50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>P</td>\n",
       "      <td>P463S</td>\n",
       "      <td>463</td>\n",
       "      <td>S</td>\n",
       "      <td>142.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.000062</td>\n",
       "      <td>delta_expr</td>\n",
       "      <td>-1.24240</td>\n",
       "      <td>&gt;=50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>S</td>\n",
       "      <td>S494L</td>\n",
       "      <td>494</td>\n",
       "      <td>L</td>\n",
       "      <td>144.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>0.000062</td>\n",
       "      <td>delta_expr</td>\n",
       "      <td>-1.15532</td>\n",
       "      <td>&gt;=50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Y</td>\n",
       "      <td>Y508C</td>\n",
       "      <td>508</td>\n",
       "      <td>C</td>\n",
       "      <td>51.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.000022</td>\n",
       "      <td>delta_expr</td>\n",
       "      <td>-1.05878</td>\n",
       "      <td>&gt;=50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Q</td>\n",
       "      <td>Q414H</td>\n",
       "      <td>414</td>\n",
       "      <td>H</td>\n",
       "      <td>308.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.000134</td>\n",
       "      <td>delta_expr</td>\n",
       "      <td>-1.05292</td>\n",
       "      <td>&gt;=50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>A</td>\n",
       "      <td>A348T</td>\n",
       "      <td>348</td>\n",
       "      <td>T</td>\n",
       "      <td>84.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>0.000036</td>\n",
       "      <td>delta_expr</td>\n",
       "      <td>-1.04997</td>\n",
       "      <td>&gt;=50</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for metric in ['bind', 'expr']:\n",
    "    m=f\"delta_{metric}\"\n",
    "    score_filter=new_thresholds[m]\n",
    "    print(f'Mutations with >=50 GISAID counts but with {metric} score < {score_filter}')\n",
    "    display(HTML(dms_scores\n",
    "                 .query('metric==@m & count >= 50 & score < @score_filter')\n",
    "                 .drop_duplicates()\n",
    "                 .sort_values(by='score')\n",
    "                 .head(20)\n",
    "                 .to_html(index=False)\n",
    "                )\n",
    "           )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Here are the scores for mutations to disulfide bonds:\n",
      "Saving plot to results/bind_expr_filters/disulfide-histogram.pdf\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x250 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Here are the scores for mutations to disulfide bonds:')\n",
    "\n",
    "p = (ggplot(dms_scores\n",
    "            .replace({'delta_bind':'ACE2 binding', 'delta_expr':'RBD expression'})\n",
    "            .assign(wildtype=lambda x: x['mutation'].str[0])\n",
    "            .query('wildtype==\"C\" & mutant!=\"C\"')\n",
    "           ) +\n",
    "     aes(x='score') + \n",
    "     geom_histogram(binwidth=0.25) +\n",
    "     geom_vline(data=new_filter_df.replace({'delta_bind':'ACE2 binding', 'delta_expr':'RBD expression'}),\n",
    "                     mapping=aes(xintercept='score'),\n",
    "                    linetype='dashed',\n",
    "                    color=CBPALETTE[1])+\n",
    "     facet_wrap('~ metric') +\n",
    "     theme_classic() +\n",
    "     theme(figure_size=(2.5 * 2, 2.5 * 1),\n",
    "           plot_title=element_text(size=14),\n",
    "           axis_text_x=element_text(angle=90),\n",
    "           strip_background=element_blank(),\n",
    "           strip_text=element_text(size=12),\n",
    "          ) +\n",
    "     xlab('B.1.351 RBD DMS score')\n",
    "     )\n",
    "\n",
    "fig = p.draw()\n",
    "\n",
    "plotfile = os.path.join(resultsdir, f\"disulfide-histogram.pdf\")\n",
    "print(f\"Saving plot to {plotfile}\")\n",
    "p.save(plotfile, verbose=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Get the bind and expr scores that correspond to the 5th percentile of mutations observed at least 50x in GISAID"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>count_threshold</th>\n",
       "      <th>percentile</th>\n",
       "      <th>bind_count</th>\n",
       "      <th>expr_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>50</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.851553</td>\n",
       "      <td>-1.991598</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50</td>\n",
       "      <td>2.5</td>\n",
       "      <td>-1.141998</td>\n",
       "      <td>-1.474578</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50</td>\n",
       "      <td>5.0</td>\n",
       "      <td>-0.831732</td>\n",
       "      <td>-1.054092</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50</td>\n",
       "      <td>10.0</td>\n",
       "      <td>-0.600498</td>\n",
       "      <td>-0.680866</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50</td>\n",
       "      <td>25.0</td>\n",
       "      <td>-0.210150</td>\n",
       "      <td>-0.276720</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def get_filter(scores_df, metric, count_threshold, percentile):\n",
    "    \n",
    "    scores=(scores_df\n",
    "            .query('metric==@metric & count >=@count_threshold')\n",
    "            .dropna()\n",
    "            )['score'].tolist()\n",
    "            \n",
    "    c=np.percentile(scores, percentile)\n",
    "    \n",
    "    return c\n",
    "\n",
    "count_thresholds = [50]\n",
    "percentiles=[1,2.5,5,10,25]\n",
    "\n",
    "v=[]\n",
    "\n",
    "for i in count_thresholds:\n",
    "    for p in percentiles:\n",
    "        t=(i,p)\n",
    "        \n",
    "        scores=(dms_scores)\n",
    "        bind_filter=get_filter(scores, 'delta_bind', i, p)\n",
    "        expr_filter=get_filter(scores, 'delta_expr', i, p)\n",
    "        \n",
    "        t=(i, p, bind_filter, expr_filter)\n",
    "        \n",
    "        v.append(t)\n",
    "        \n",
    "\n",
    "df = pd.DataFrame(v, columns =['count_threshold', 'percentile', 'bind_count', 'expr_count'])\n",
    "display(HTML(df.to_html(index=False)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>site</th>\n",
       "      <th>mutant</th>\n",
       "      <th>delta_bind</th>\n",
       "      <th>delta_expr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>331</td>\n",
       "      <td>A</td>\n",
       "      <td>-0.03</td>\n",
       "      <td>-0.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>331</td>\n",
       "      <td>C</td>\n",
       "      <td>-0.09</td>\n",
       "      <td>-1.26</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "og_dms_scores=(pd.read_csv(og_dms_file)\n",
    "               # remove extraneous columns\n",
    "               .drop(columns=['site_RBD','wildtype', 'mutation', 'mutation_RBD', 'bind_lib1', 'bind_lib2', 'expr_lib1', 'expr_lib2'])\n",
    "               # rename some columns\n",
    "               .rename(columns={'site_SARS2':'site', 'bind_avg':'delta_bind', 'expr_avg':'delta_expr'})\n",
    "              )\n",
    "\n",
    "display(HTML(og_dms_scores.head(2).to_html(index=False)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>wildtype</th>\n",
       "      <th>mutation</th>\n",
       "      <th>site</th>\n",
       "      <th>mutant</th>\n",
       "      <th>count</th>\n",
       "      <th>n_countries</th>\n",
       "      <th>frequency</th>\n",
       "      <th>metric</th>\n",
       "      <th>score</th>\n",
       "      <th>count_categories</th>\n",
       "      <th>wuhan1dms_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331A</td>\n",
       "      <td>331</td>\n",
       "      <td>A</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>0.12426</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331C</td>\n",
       "      <td>331</td>\n",
       "      <td>C</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>-0.20203</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331D</td>\n",
       "      <td>331</td>\n",
       "      <td>D</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>8.673362e-07</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>0.00131</td>\n",
       "      <td>1 to 9</td>\n",
       "      <td>0.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331E</td>\n",
       "      <td>331</td>\n",
       "      <td>E</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>0.10980</td>\n",
       "      <td>0</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>N</td>\n",
       "      <td>N331F</td>\n",
       "      <td>331</td>\n",
       "      <td>F</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>delta_bind</td>\n",
       "      <td>-0.15674</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dms_scores=(dms_scores\n",
    "            .merge((og_dms_scores\n",
    "                    .melt(id_vars=['site', 'mutant',],\n",
    "                          value_vars=['delta_bind', 'delta_expr'], \n",
    "                          var_name='metric', \n",
    "                          value_name='wuhan1dms_score',\n",
    "                         )\n",
    "                   ),\n",
    "                   how='left',\n",
    "                   on=['site', 'mutant', 'metric'],\n",
    "                   validate='many_to_one'\n",
    "                  )\n",
    "           )\n",
    "display(HTML(dms_scores.head().to_html(index=False)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/fh/fast/bloom_j/computational_notebooks/agreaney/2021/SARS-CoV-2-RBD_B.1.351/env/lib/python3.8/site-packages/plotnine/layer.py:401: PlotnineWarning: geom_point : Removed 72 rows containing missing values.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving plot to results/bind_expr_filters/wuhan-b1351-corr.pdf\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/fh/fast/bloom_j/computational_notebooks/agreaney/2021/SARS-CoV-2-RBD_B.1.351/env/lib/python3.8/site-packages/plotnine/layer.py:401: PlotnineWarning: geom_point : Removed 72 rows containing missing values.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x250 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dms_score_corrs = (dms_scores\n",
    "                   .replace({'delta_bind':'ACE2 binding', 'delta_expr':'RBD expression'})\n",
    "                   .assign(is_disulfide=lambda x: x['mutation'].apply(lambda s: s[0]).isin([\"C\"]))\n",
    "                  )\n",
    "\n",
    "# plot correlations\n",
    "xmin = dms_score_corrs['wuhan1dms_score'].min()\n",
    "xspan = dms_score_corrs['wuhan1dms_score'].max() - xmin\n",
    "ymin = dms_score_corrs['score'].min()\n",
    "yspan = dms_score_corrs['score'].max() - ymin\n",
    "\n",
    "p = (ggplot(dms_score_corrs) +\n",
    "     aes('wuhan1dms_score', 'score') +\n",
    "     geom_point(aes(fill='is_disulfide', alpha='is_disulfide'), color='black') +\n",
    "     geom_vline(data=og_thresholds_df.replace({'delta_bind':'ACE2 binding', 'delta_expr':'RBD expression'}),\n",
    "                     mapping=aes(xintercept='score'),\n",
    "                    linetype='dashed',\n",
    "                    color=CBPALETTE[1])+\n",
    "     geom_hline(data=new_filter_df.replace({'delta_bind':'ACE2 binding', 'delta_expr':'RBD expression'}),\n",
    "                     mapping=aes(yintercept='score'),\n",
    "                    linetype='dashed',\n",
    "                    color=CBPALETTE[1])+\n",
    "     facet_wrap('~ metric') +\n",
    "     theme_classic() +\n",
    "     theme(figure_size=(2.5 * 2, 2.5 * 1),\n",
    "           plot_title=element_text(size=14),\n",
    "           strip_background=element_blank(),\n",
    "           strip_text=element_text(size=12),\n",
    "          ) +\n",
    "     xlab('Wuhan-Hu-1 DMS score from Starr et al. (2020)') +\n",
    "     ylab('B.1.351 RBD DMS score') +\n",
    "     scale_fill_manual(values=['black', 'red']) +\n",
    "     scale_alpha_manual(values=[0.2,1]) +\n",
    "     labs(fill = 'mutation to disulfide bond?',\n",
    "          alpha = 'mutation to disulfide bond?'\n",
    "         )\n",
    "     )\n",
    "\n",
    "p.draw()\n",
    "plotfile = os.path.join(resultsdir, f\"wuhan-b1351-corr.pdf\")\n",
    "print(f\"Saving plot to {plotfile}\")\n",
    "p.save(plotfile, verbose=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mutations from the Wuhan-Hu-1 library that:\n",
      "pass bind: 3422\n",
      "pass expr: 2328\n",
      "pass both: 2269\n"
     ]
    }
   ],
   "source": [
    "print('Mutations from the Wuhan-Hu-1 library that:')\n",
    "print('pass bind: '+ str(len(og_dms_scores.query('delta_bind >= -2.35'))))\n",
    "print('pass expr: '+ str(len(og_dms_scores.query('delta_expr >= -1.0'))))\n",
    "print('pass both: '+ str(len(og_dms_scores.query('delta_bind >= -2.35 & delta_expr >= -1.0'))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "B.1.351 mutations that \n",
      "pass bind: 3291 \n",
      "pass expr: 2369 \n",
      "pass both: 2266 \n",
      "pass both and not disulfide: 2207\n",
      "Pass bind, expr, not disulfide, and not WT: 2014\n",
      "Total number of possible mutations to non-disulfide sites: 3653\n"
     ]
    }
   ],
   "source": [
    "bind_threshold=new_thresholds['delta_bind']\n",
    "expr_threshold=new_thresholds['delta_expr']\n",
    "        \n",
    "n_bind=len(dms_scores.query('metric==\"delta_bind\" & score >= @bind_threshold'))\n",
    "n_expr=len(dms_scores.query('metric==\"delta_expr\" & score >= @expr_threshold'))\n",
    "\n",
    "df=(dms_scores\n",
    "     .pivot_table(index=['mutation', 'wildtype', 'mutant'],\n",
    "                  values=['score'],\n",
    "                  columns=['metric'],\n",
    "                 )\n",
    "     .reset_index()\n",
    "       )\n",
    "\n",
    "df.columns=['mutation', 'wildtype', 'mutant','delta_bind', 'delta_expr']\n",
    "\n",
    "n_both=len(df\n",
    "           .query('delta_bind >= @bind_threshold & delta_expr >= @expr_threshold')\n",
    "          )\n",
    "        \n",
    "n_both_notC=len((df\n",
    "                .assign(not_disulfide=lambda x: x['mutation'].str[0] != \"C\")\n",
    "                .query('delta_bind >= @bind_threshold & delta_expr >= @expr_threshold & not_disulfide')\n",
    "          ))\n",
    "\n",
    "n_both_notC_notWT=len((df\n",
    "                .assign(not_disulfide=lambda x: x['mutation'].str[0] != \"C\")\n",
    "                .assign(not_WT=lambda x: x['wildtype']!=x['mutant'])\n",
    "                .query('delta_bind >= @bind_threshold & delta_expr >= @expr_threshold & not_disulfide & not_WT')\n",
    "          ))\n",
    "\n",
    "total_muts_notC=len((df\n",
    "                .assign(not_disulfide=lambda x: x['mutation'].str[0] != \"C\")\n",
    "                .assign(not_WT=lambda x: x['wildtype']!=x['mutant'])\n",
    "                .query('not_disulfide & not_WT')\n",
    "          ))\n",
    "\n",
    "print(f'B.1.351 mutations that \\npass bind: {n_bind} \\npass expr: {n_expr} \\npass both: {n_both} \\npass both and not disulfide: {n_both_notC}')\n",
    "print(f'Pass bind, expr, not disulfide, and not WT: {n_both_notC_notWT}')\n",
    "\n",
    "print(f'Total number of possible mutations to non-disulfide sites: {total_muts_notC}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This percentage of all variants seen >=50x in GISAID are retained by the binding filter of -3.0\n",
      "99.4\n",
      "This percentage of all variants seen >=50x in GISAID are retained by the expression filter of -1.0\n",
      "93.8\n"
     ]
    }
   ],
   "source": [
    "print(f'This percentage of all variants seen >=50x in GISAID are retained by the binding filter of {bind_threshold}')\n",
    "print(round(100-stats.percentileofscore((dms_scores\n",
    "                               .query('metric==\"delta_bind\" & count>=50')['score']), \n",
    "                              bind_threshold, \n",
    "                              kind='rank'\n",
    "                             ),\n",
    "            1\n",
    "           )\n",
    "     )\n",
    "\n",
    "print(f'This percentage of all variants seen >=50x in GISAID are retained by the expression filter of {expr_threshold}')\n",
    "print(round(100-stats.percentileofscore((dms_scores\n",
    "                               .query('metric==\"delta_expr\" & count>=50')['score']), \n",
    "                              expr_threshold, \n",
    "                              kind='rank'\n",
    "                             ),\n",
    "            1\n",
    "           )\n",
    "     )\n",
    "\n",
    "\n",
    "# dms_scores.query('metric==\"delta_bind\" & score >= @bind_threshold & count>=50')['score'].min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "language_info": {
   "name": "python"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}