07_Suspect_Actives.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Suspect Actives\n",
    "\n",
    "Inspect cases where the number of inactives for a compound is greater than the number of actives."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from collections import defaultdict\n",
    "\n",
    "from math import log10, floor, ceil\n",
    "\n",
    "from itertools import chain\n",
    "\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configuration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "marker_class_col, color_class_col = 'standard_relation', 'description'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# ChEMBL connection...\n",
    "\n",
    "engine = create_engine(open('database.txt').read().strip())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "sns.set()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Get suspect actives\n",
    "\n",
    "See SQL files '[tt_activity_counts](SQL/tt_activity_counts.sql)' for further details."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------------------------------------------------------------\n",
      "-- \n",
      "-- Count active and inactive records for compounds.\n",
      "-- \n",
      "-- Note that 'compounds' are here defined in terms of USMILES.\n",
      "-- \n",
      "-- Note also that 'targets' are defined in terms of symbol/species pairs.\n",
      "--\n",
      "-- Note use of source table 'tt_curve_data_v2'; this is to get a full picture of the inactives.\n",
      "-- \n",
      "----------------------------------------------------------------------------------------------------\n",
      "\n",
      "-- drop table tt_activity_counts;\n",
      "\n",
      "--\n",
      "\n",
      "-- \n",
      "\n",
      "create table tt_activity_counts as\n",
      "select\n",
      "    a.*\n",
      "  , case when ((a.n_active > 0) and (a.n_inactive > a.n_active)) then 1 else 0 end as suspect\n",
      "from (\n",
      "  select\n",
      "      a.symbol\n",
      "    , a.species\n",
      "    , b.usmiles\n",
      "    , count(case when nvl(a.active, 0) = 1 then 1 end) n_active\n",
      "    , count(case when nvl(a.active, 0) = 0 then 1 end) n_inactive\n",
      "    , count(*) n_total\n",
      "  from\n",
      "    tt_curve_data_v2 a\n",
      "    join tt_structure_lookup b on a.parent_cmpd_chemblid = b.cmpd\n",
      "  where\n",
      "    a.exclude = 0\n",
      "  group by\n",
      "      a.symbol\n",
      "    , a.species\n",
      "    , b.usmiles\n",
      "  ) a\n",
      "order by\n",
      "    a.symbol\n",
      "  , a.species\n",
      "  , a.usmiles\n",
      ";\n",
      "\n",
      "----------------------------------------------------------------------------------------------------\n",
      "-- End\n",
      "----------------------------------------------------------------------------------------------------\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(open('SQL/tt_activity_counts.sql').read())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(270, 5)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# NB Complicated query is to avoid division-by-zero errors...\n",
    "\n",
    "suspect = pd.read_sql(\"\"\"\n",
    "select \n",
    "    a.symbol\n",
    "  , a.species\n",
    "  , count(case when a.suspect = 1 then 1 end) as n_suspect\n",
    "  , count(case when a.n_active > 0 then 1 end) as n_active\n",
    "  , round(count(case when a.suspect = 1 then 1 end)/count(case when a.n_active > 0 then 1 end) * 100.0, 1) as pct_suspect\n",
    "from\n",
    "  tt_activity_counts a\n",
    "  join (\n",
    "    select \n",
    "        symbol\n",
    "      , species\n",
    "      , count(case when n_active > 0 then 1 end) as n_active\n",
    "    from\n",
    "      tt_activity_counts a\n",
    "    group by\n",
    "        symbol\n",
    "      , species\n",
    "  ) b on a.symbol = b.symbol and a.species = b.species\n",
    "where\n",
    "  b.n_active > 0\n",
    "group by\n",
    "    a.symbol\n",
    "  , a.species\n",
    "order by\n",
    "  pct_suspect desc\n",
    "  , a.symbol\n",
    "  , a.species\n",
    "\"\"\", engine)\n",
    "\n",
    "suspect.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>symbol</th>\n",
       "      <th>species</th>\n",
       "      <th>n_suspect</th>\n",
       "      <th>n_active</th>\n",
       "      <th>pct_suspect</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>GRIN1</td>\n",
       "      <td>Rat</td>\n",
       "      <td>108</td>\n",
       "      <td>706</td>\n",
       "      <td>15.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>KCNJ11</td>\n",
       "      <td>Human</td>\n",
       "      <td>8</td>\n",
       "      <td>84</td>\n",
       "      <td>9.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>VDR</td>\n",
       "      <td>Human</td>\n",
       "      <td>12</td>\n",
       "      <td>176</td>\n",
       "      <td>6.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>GRIN1</td>\n",
       "      <td>Human</td>\n",
       "      <td>17</td>\n",
       "      <td>273</td>\n",
       "      <td>6.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>AHR</td>\n",
       "      <td>Human</td>\n",
       "      <td>6</td>\n",
       "      <td>172</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>CHRNA1</td>\n",
       "      <td>Human</td>\n",
       "      <td>2</td>\n",
       "      <td>58</td>\n",
       "      <td>3.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>DRD1</td>\n",
       "      <td>Human</td>\n",
       "      <td>23</td>\n",
       "      <td>687</td>\n",
       "      <td>3.3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    symbol species  n_suspect  n_active  pct_suspect\n",
       "0    GRIN1     Rat        108       706         15.3\n",
       "3   KCNJ11   Human          8        84          9.5\n",
       "5      VDR   Human         12       176          6.8\n",
       "7    GRIN1   Human         17       273          6.2\n",
       "10     AHR   Human          6       172          3.5\n",
       "11  CHRNA1   Human          2        58          3.4\n",
       "12    DRD1   Human         23       687          3.3"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "suspect.query(\"n_active > 50 and pct_suspect >= 3\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Targets\n",
    "\n",
    "Inspect data for targets with high fraction of 'suspect' data points."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# symbol, species = 'GRIN1', 'Rat'\n",
    "\n",
    "symbol, species = 'KCNJ11', 'Human'\n",
    "\n",
    "# symbol, species = 'VDR', 'Human'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Minimum number of activities to keep assay in plot..\n",
    "\n",
    "min_n = 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(34, 36)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get raw data...\n",
    "\n",
    "data = pd.read_sql(\"\"\"\n",
    "select\n",
    "      a.*\n",
    "    , b.usmiles\n",
    "from\n",
    "     tt_curve_data_v2 a\n",
    "join tt_structure_lookup b on a.parent_cmpd_chemblid = b.cmpd\n",
    "join tt_activity_counts c on a.symbol = c.symbol and a.species = c.species and b.usmiles = c.usmiles\n",
    "where\n",
    "    a.symbol  = :symbol\n",
    "and a.species = :species\n",
    "and c.suspect = 1\n",
    "and a.exclude = 0\n",
    "and a.nat >= 10\n",
    "and a.nat <= 50\n",
    "order by\n",
    "    b.usmiles\n",
    "  , a.parent_cmpd_chemblid\n",
    "  , a.description\n",
    "  , a.standard_relation\n",
    "  , a.active\n",
    "\"\"\", engine, params={'symbol': symbol, 'species': species})\n",
    "\n",
    "data.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Table of assays"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(13, 3)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Table of assays...\n",
    "\n",
    "table_of_assays = data.groupby(['description', 'assay_chemblid']).apply(len).to_frame(name='n').reset_index().sort(['n', 'description'], ascending=[0, 1])\n",
    "\n",
    "table_of_assays.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(9, 3)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Discard smaller assays...\n",
    "\n",
    "table_of_assays.query(\"n < @min_n\").shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4, 3)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "table_of_assays = table_of_assays.query(\"n >= @min_n\")\n",
    "\n",
    "table_of_assays.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(25, 36)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Take only data from larger assays...\n",
    "\n",
    "assay_chemblids = table_of_assays['assay_chemblid'].tolist()\n",
    "\n",
    "data_0 = data\n",
    "\n",
    "data = data.query(\"assay_chemblid in @assay_chemblids\").reset_index(drop=True)\n",
    "\n",
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Add index number of USMILES (i.e. as they will appear on X-axis of plot)...\n",
    "\n",
    "data['usmiles_n'] = pd.Categorical.from_array(data['usmiles']).categories.get_indexer(data['usmiles'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Table of compounds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8, 2)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Table of compounds...\n",
    "\n",
    "table_of_compounds = data[['usmiles', 'parent_cmpd_chemblid']].drop_duplicates().groupby('usmiles').apply(lambda x: sorted(x['parent_cmpd_chemblid'])).to_frame('cmpds').reset_index()\n",
    "\n",
    "table_of_compounds.index.name = 'usmiles_n'\n",
    "\n",
    "table_of_compounds.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plot chart"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "data[marker_class_col] = data[marker_class_col].fillna(' ')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Colours...\n",
    "\n",
    "def get_color_map(n_classes):\n",
    "\n",
    "    n1 = int(round(n_classes/2))*2 # Round up to next even number for convenience\n",
    "        \n",
    "    color_map = plt.get_cmap('jet')(np.linspace(0.0, 1.0, n1))\n",
    "\n",
    "    return color_map[list(chain.from_iterable(zip(range(0, int(n1/2)), range(int(n1/2), n1))))[:n_classes]]\n",
    "\n",
    "color_classes = sorted(data[color_class_col].unique())\n",
    "\n",
    "color_map = get_color_map(len(color_classes))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Marker shape...\n",
    "\n",
    "marker_classes = sorted(data[marker_class_col].unique())\n",
    "\n",
    "marker_list = ['o', 's', '^']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# List of USMILES for X-axis...\n",
    "\n",
    "usmiles = table_of_compounds['usmiles'].tolist()\n",
    "\n",
    "len(usmiles)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAACOEAAAKpCAYAAADw5KoXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nJEmSJEmSJEmSJM3TfwHThPY9GD6BdAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11a857910>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot...\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "\n",
    "fig.set_size_inches(40, 10)\n",
    "\n",
    "# X-axis...\n",
    "\n",
    "x_positions = range(len(usmiles))\n",
    "\n",
    "ax.set_xlim((x_positions[0]-0.5, x_positions[-1]+0.5))\n",
    "\n",
    "ax.set_xticks(x_positions)\n",
    "\n",
    "ax.tick_params(axis='both', direction='out')\n",
    "\n",
    "ax.set_xticklabels(x_positions, fontsize=10, rotation=90)\n",
    "\n",
    "ax.set_xlabel('USMILES number')\n",
    "\n",
    "# Y-axis...\n",
    "\n",
    "y = [-log10(x/10**9) if not np.isnan(x) else 3.0 for x in data['standard_value']]\n",
    "\n",
    "ax.set_ylim((floor(min(y))-0.25, ceil(max(y))+0.25))\n",
    "\n",
    "ax.set_ylabel('pActivity')\n",
    "            \n",
    "# Add line...\n",
    "            \n",
    "ax.plot((x_positions[0]-0.5, x_positions[-1]+0.5), (5, 5), color='m', linestyle='--', linewidth=2)\n",
    "\n",
    "######\n",
    "\n",
    "point_sets = defaultdict(list)\n",
    "\n",
    "for i, color_class in enumerate(color_classes):\n",
    "    \n",
    "    ### logging.info(\"{}) color_class: '{}'\".format(i, color_class))\n",
    "        \n",
    "    for j, marker_class in enumerate(marker_classes):\n",
    "        \n",
    "        ### logging.info(\"  {}) marker_class: '{}'\".format(j, marker_class))\n",
    "        \n",
    "        subset = data.query(\"({} == @color_class) and ({} == @marker_class)\".format(color_class_col, marker_class_col))\n",
    "        \n",
    "        x_subset = [usmiles.index(x) for x in subset['usmiles']]\n",
    "        \n",
    "        ### logging.info(\"    x_subset: [{}]\".format(', '.join(\"{}\".format(x) for x in x_subset)))\n",
    "        \n",
    "        y_subset = [-log10(x/10**9) if not np.isnan(x) else 3.0 for x in subset['standard_value']]\n",
    "        \n",
    "        assert(len(x_subset) == len(y_subset))\n",
    "                \n",
    "        ### logging.info(\"    y_subset: [{}]\".format(', '.join(\"{}\".format(round(x, 2)) for x in y_subset)))\n",
    "        \n",
    "        point_set = ax.scatter(x_subset, y_subset, s=100, color=color_map[color_classes.index(color_class)], marker=marker_list[marker_classes.index(marker_class)], label=\"{}, {}\".format(color_class, marker_class))\n",
    "        \n",
    "        point_sets[i].append(point_set)\n",
    "                \n",
    "######\n",
    "\n",
    "legend0 = ax.legend(point_sets[0], marker_classes, bbox_to_anchor=(0.0, -0.1), loc=2)\n",
    "\n",
    "legend1 = ax.legend([x[0] for x in point_sets.values()], color_classes, bbox_to_anchor=(0.05, -0.1), loc=2)\n",
    "\n",
    "_ = ax.add_artist(legend0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "_N.B._ 3 is used as a dummy-value for missing activity values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>usmiles</th>\n",
       "      <th>cmpds</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>usmiles_n</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CC(C)(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "      <td>[CHEMBL336586]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CC(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "      <td>[CHEMBL131971]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CC1(NC2=NS(=O)(=O)c3sc(Cl)cc3N2)CCC1</td>\n",
       "      <td>[CHEMBL134168]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CC1=NS(=O)(=O)c2cc(Cl)ccc2N1</td>\n",
       "      <td>[CHEMBL181]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CCC(C)(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "      <td>[CHEMBL135091]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>CCC(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "      <td>[CHEMBL2111887]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>CCCNC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "      <td>[CHEMBL133596]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>O=S1(=O)NC(NC2CCC2)=Nc2cc(Cl)sc21</td>\n",
       "      <td>[CHEMBL133663]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Inspect compounds...\n",
    "\n",
    "HTML(table_of_compounds.to_html())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>assay_chemblid</th>\n",
       "      <th>description</th>\n",
       "      <th>parent_cmpd_chemblid</th>\n",
       "      <th>standard_relation</th>\n",
       "      <th>standard_value</th>\n",
       "      <th>standard_units</th>\n",
       "      <th>pchembl_value</th>\n",
       "      <th>active</th>\n",
       "      <th>usmiles_n</th>\n",
       "      <th>usmiles</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CHEMBL810541</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at high affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL336586</td>\n",
       "      <td>=</td>\n",
       "      <td>230</td>\n",
       "      <td>nM</td>\n",
       "      <td>6.64</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>CC(C)(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CHEMBL810542</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at low affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL336586</td>\n",
       "      <td>=</td>\n",
       "      <td>393000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>CC(C)(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CHEMBL810543</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 without ATP</td>\n",
       "      <td>CHEMBL336586</td>\n",
       "      <td>=</td>\n",
       "      <td>1350000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>CC(C)(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CHEMBL810541</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at high affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL131971</td>\n",
       "      <td>=</td>\n",
       "      <td>340</td>\n",
       "      <td>nM</td>\n",
       "      <td>6.47</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>CC(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CHEMBL810542</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at low affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL131971</td>\n",
       "      <td>=</td>\n",
       "      <td>187000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>CC(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>CHEMBL810543</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 without ATP</td>\n",
       "      <td>CHEMBL131971</td>\n",
       "      <td>=</td>\n",
       "      <td>554000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>CC(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>CHEMBL810541</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at high affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL134168</td>\n",
       "      <td>=</td>\n",
       "      <td>3</td>\n",
       "      <td>nM</td>\n",
       "      <td>8.52</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>CC1(NC2=NS(=O)(=O)c3sc(Cl)cc3N2)CCC1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>CHEMBL810542</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at low affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL134168</td>\n",
       "      <td>=</td>\n",
       "      <td>10800</td>\n",
       "      <td>nM</td>\n",
       "      <td>4.97</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>CC1(NC2=NS(=O)(=O)c3sc(Cl)cc3N2)CCC1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>CHEMBL810543</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 without ATP</td>\n",
       "      <td>CHEMBL134168</td>\n",
       "      <td>=</td>\n",
       "      <td>859000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>CC1(NC2=NS(=O)(=O)c3sc(Cl)cc3N2)CCC1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>CHEMBL810543</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 without ATP</td>\n",
       "      <td>CHEMBL181</td>\n",
       "      <td>=</td>\n",
       "      <td>750000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>CC1=NS(=O)(=O)c2cc(Cl)ccc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>CHEMBL810541</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at high affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL135091</td>\n",
       "      <td>=</td>\n",
       "      <td>180</td>\n",
       "      <td>nM</td>\n",
       "      <td>6.74</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>CCC(C)(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>CHEMBL810542</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at low affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL135091</td>\n",
       "      <td>=</td>\n",
       "      <td>375000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>CCC(C)(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>CHEMBL810543</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 without ATP</td>\n",
       "      <td>CHEMBL135091</td>\n",
       "      <td>=</td>\n",
       "      <td>899000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>CCC(C)(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>CHEMBL810541</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at high affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL2111887</td>\n",
       "      <td>=</td>\n",
       "      <td>460</td>\n",
       "      <td>nM</td>\n",
       "      <td>6.34</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>CCC(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>CHEMBL810541</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at high affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL2111887</td>\n",
       "      <td>=</td>\n",
       "      <td>640</td>\n",
       "      <td>nM</td>\n",
       "      <td>6.19</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>CCC(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>CHEMBL810542</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at low affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL2111887</td>\n",
       "      <td>=</td>\n",
       "      <td>224000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>CCC(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>CHEMBL810542</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at low affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL2111887</td>\n",
       "      <td>=</td>\n",
       "      <td>309000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>CCC(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>CHEMBL810540</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 without ATP</td>\n",
       "      <td>CHEMBL2111887</td>\n",
       "      <td>=</td>\n",
       "      <td>487000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>CCC(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>CHEMBL810540</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 without ATP</td>\n",
       "      <td>CHEMBL2111887</td>\n",
       "      <td>=</td>\n",
       "      <td>420000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>CCC(C)NC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>CHEMBL810541</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at high affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL133596</td>\n",
       "      <td>=</td>\n",
       "      <td>2270</td>\n",
       "      <td>nM</td>\n",
       "      <td>5.64</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>CCCNC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>CHEMBL810542</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at low affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL133596</td>\n",
       "      <td>=</td>\n",
       "      <td>584000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>CCCNC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>CHEMBL810543</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 without ATP</td>\n",
       "      <td>CHEMBL133596</td>\n",
       "      <td>=</td>\n",
       "      <td>924000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>CCCNC1=NS(=O)(=O)c2sc(Cl)cc2N1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>CHEMBL810541</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at high affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL133663</td>\n",
       "      <td>=</td>\n",
       "      <td>120</td>\n",
       "      <td>nM</td>\n",
       "      <td>6.92</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>O=S1(=O)NC(NC2CCC2)=Nc2cc(Cl)sc21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>CHEMBL810542</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at low affinity state with 2 mM MgATP</td>\n",
       "      <td>CHEMBL133663</td>\n",
       "      <td>=</td>\n",
       "      <td>146000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>O=S1(=O)NC(NC2CCC2)=Nc2cc(Cl)sc21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>CHEMBL810543</td>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 without ATP</td>\n",
       "      <td>CHEMBL133663</td>\n",
       "      <td>=</td>\n",
       "      <td>943000</td>\n",
       "      <td>nM</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>O=S1(=O)NC(NC2CCC2)=Nc2cc(Cl)sc21</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Inspect raw data...\n",
    "    \n",
    "cols = ['assay_chemblid', 'description', 'parent_cmpd_chemblid', 'standard_relation', 'standard_value', 'standard_units', 'pchembl_value', 'active', 'usmiles_n', 'usmiles']\n",
    "\n",
    "HTML(data[cols].to_html())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>description</th>\n",
       "      <th>assay_chemblid</th>\n",
       "      <th>n</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at high affinity state with 2 mM MgATP</td>\n",
       "      <td><a target=\"_blank\" href=\"https://www.ebi.ac.uk/chembl/assay/inspect/CHEMBL810541\">CHEMBL810541</a></td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at low affinity state with 2 mM MgATP</td>\n",
       "      <td><a target=\"_blank\" href=\"https://www.ebi.ac.uk/chembl/assay/inspect/CHEMBL810542\">CHEMBL810542</a></td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 without ATP</td>\n",
       "      <td><a target=\"_blank\" href=\"https://www.ebi.ac.uk/chembl/assay/inspect/CHEMBL810543\">CHEMBL810543</a></td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 without ATP</td>\n",
       "      <td><a target=\"_blank\" href=\"https://www.ebi.ac.uk/chembl/assay/inspect/CHEMBL810540\">CHEMBL810540</a></td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                                                                                                            description                                                                                      assay_chemblid  n\n",
       "0  Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at high affinity state with 2 mM MgATP  <a target=\"_blank\" href=\"https://www.ebi.ac.uk/chembl/assay/inspect/CHEMBL810541\">CHEMBL810541</a>  8\n",
       "1   Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 at low affinity state with 2 mM MgATP  <a target=\"_blank\" href=\"https://www.ebi.ac.uk/chembl/assay/inspect/CHEMBL810542\">CHEMBL810542</a>  8\n",
       "2                             Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 without ATP  <a target=\"_blank\" href=\"https://www.ebi.ac.uk/chembl/assay/inspect/CHEMBL810543\">CHEMBL810543</a>  7\n",
       "3                             Inhibition of [3H]glibenclamide binding to HEK293 cells co-expressing human Sulfonylurea receptor SUR1 and Inward rectifier K+ channel Kir6.2 without ATP  <a target=\"_blank\" href=\"https://www.ebi.ac.uk/chembl/assay/inspect/CHEMBL810540\">CHEMBL810540</a>  2"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Inspect assays...\n",
    "\n",
    "def f(x):\n",
    "    \n",
    "    x['assay_chemblid'] = '<a target=\"_blank\" href=\"https://www.ebi.ac.uk/chembl/assay/inspect/{}\">{}</a>'.format(* [x['assay_chemblid']]*2)\n",
    "    \n",
    "    return x\n",
    "\n",
    "table_of_assays.sort('description').reset_index(drop=True).apply(f, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conclusions\n",
    "\n",
    "* GRIN1 / Rat: there seems to be considerable discepancies among the various assays, and merging data for the different isoforms may not be supportable in this case. This can also be seen in the notebook '[GRIN1](Correlation_Targets_Rat.html)'."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}