{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from rdkit import Chem\n",
      "from rdkit.Chem import AllChem\n",
      "from rdkit import DataStructs\n",
      "\n",
      "m1 = Chem.MolFromSmiles('O=C1CN(N=Cc2ccc([N+](=O)[O-])o2)C(=O)N1')\n",
      "m2 = Chem.MolFromSmiles('CCCC1COC(Cn2cncn2)(c2ccc(Cl)cc2Cl)O1')\n",
      "m3 = Chem.MolFromSmiles('CCCCCC=O')\n",
      "# similar to m1\n",
      "m4 = Chem.MolFromSmiles('CCCC1COC(Cn2cncn2)(c2ccc(Cl)cc2O)O1')\n",
      "\n",
      "mols = [m1, m2, m3, m4]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 14
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "fps = [AllChem.GetMorganFingerprintAsBitVect(m, 2, nBits=1024, useFeatures=False) for m in mols]\n",
      "indx_fps = dict()\n",
      "for indx, fp in enumerate(fps):\n",
      "    indx_fps[indx] = fp"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 15
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def diversity(mols_fp, threshold):\n",
      "    diverse = []\n",
      "    similar = []\n",
      "    \n",
      "    for m1 in mols_fp:\n",
      "        m1_fp = mols_fp[m1]\n",
      "        sim_vals = []\n",
      "        for m2 in mols_fp:\n",
      "            m2_fp = mols_fp[m2]\n",
      "            # if different molecules else skip\n",
      "            if m1 != m2:\n",
      "                sim_vals.append(DataStructs.FingerprintSimilarity(m1_fp, m2_fp, metric=DataStructs.TanimotoSimilarity))\n",
      "            # end if\n",
      "        # end for\n",
      "        # Get Maximum similarity\n",
      "        sim_vals = max(sim_vals)\n",
      "        if sim_vals >= threshold:\n",
      "            similar.append(m1)\n",
      "        else:\n",
      "            diverse.append(m1)\n",
      "    # end for\n",
      "    \n",
      "    return (diverse, similar)\n",
      "                "
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 16
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "(diverse, similar) = diversity(indx_fps, 0.8)\n",
      "print (diverse, similar)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "([0, 2], [1, 3])\n"
       ]
      }
     ],
     "prompt_number": 17
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}