Commit d1208fbb7c7f8802231af440548e8ce61bbac0bf
1 parent
70ad245597
Exists in
master
Updated README and Python Code
Showing
1 changed file
with
217 additions
and
0 deletions
Show diff stats
nbi_simulation_new_for_git.ipynb
| File was created | 1 | { | |
| 2 | "cells": [ | ||
| 3 | { | ||
| 4 | "cell_type": "code", | ||
| 5 | "execution_count": 1, | ||
| 6 | "metadata": { | ||
| 7 | "collapsed": false | ||
| 8 | }, | ||
| 9 | "outputs": [], | ||
| 10 | "source": [ | ||
| 11 | "import numpy as np\n", | ||
| 12 | "import csv\n", | ||
| 13 | "import numpy.matlib\n", | ||
| 14 | "from operator import itemgetter, attrgetter\n", | ||
| 15 | "from sklearn.model_selection import KFold\n", | ||
| 16 | "from sklearn.metrics import roc_curve, auc\n", | ||
| 17 | "import matplotlib.pyplot as plt" | ||
| 18 | ] | ||
| 19 | }, | ||
| 20 | { | ||
| 21 | "cell_type": "code", | ||
| 22 | "execution_count": 2, | ||
| 23 | "metadata": { | ||
| 24 | "collapsed": false | ||
| 25 | }, | ||
| 26 | "outputs": [ | ||
| 27 | { | ||
| 28 | "name": "stdout", | ||
| 29 | "output_type": "stream", | ||
| 30 | "text": [ | ||
| 31 | "6319 3772 (3772, 6319) 26093.0 (6319, 3772)\n" | ||
| 32 | ] | ||
| 33 | } | ||
| 34 | ], | ||
| 35 | "source": [ | ||
| 36 | "#fp = open('dt_new_and_fda_unique.csv','r')\n", | ||
| 37 | "fp = open('new_dt_from_go_and_db_unique_latest.csv','r')\n", | ||
| 38 | "drugid = []\n", | ||
| 39 | "targetid = []\n", | ||
| 40 | "\n", | ||
| 41 | "##1. Reading edge list line by line##\n", | ||
| 42 | "for line in fp:\n", | ||
| 43 | " line = line.strip()\n", | ||
| 44 | " tmp = line.split(',')\n", | ||
| 45 | " drugid.append(tmp[0])\n", | ||
| 46 | " targetid.append(tmp[1])\n", | ||
| 47 | "fp.close()\n", | ||
| 48 | "##End 1##\n", | ||
| 49 | "\n", | ||
| 50 | "drug = np.array(drugid)\n", | ||
| 51 | "target =np.array(targetid)\n", | ||
| 52 | "\n", | ||
| 53 | "uni_drugid = np.unique(np.array(drugid))\n", | ||
| 54 | "uni_targetid = np.unique(np.array(targetid))\n", | ||
| 55 | "\n", | ||
| 56 | "##creating zero incidence matrix for the graph##\n", | ||
| 57 | "\n", | ||
| 58 | "A = np.zeros((uni_targetid.shape[0], uni_drugid.shape[0]))\n", | ||
| 59 | "\n", | ||
| 60 | "for i in range(len(drugid)):\n", | ||
| 61 | " idx1 = np.where(uni_targetid==targetid[i])\n", | ||
| 62 | " idx2 = np.where(uni_drugid==drugid[i])\n", | ||
| 63 | " A[idx1,idx2] = 1\n", | ||
| 64 | "\n", | ||
| 65 | "nd = uni_drugid.shape[0]\n", | ||
| 66 | "mt = uni_targetid.shape[0]\n", | ||
| 67 | "\n", | ||
| 68 | "A_T = np.transpose(A)\n", | ||
| 69 | "no_edges = np.sum(A)\n", | ||
| 70 | "print nd, mt, A.shape, np.sum(A), A_T.shape\n" | ||
| 71 | ] | ||
| 72 | }, | ||
| 73 | { | ||
| 74 | "cell_type": "code", | ||
| 75 | "execution_count": 3, | ||
| 76 | "metadata": { | ||
| 77 | "collapsed": false | ||
| 78 | }, | ||
| 79 | "outputs": [ | ||
| 80 | { | ||
| 81 | "name": "stdout", | ||
| 82 | "output_type": "stream", | ||
| 83 | "text": [ | ||
| 84 | "(3772, 3772)\n", | ||
| 85 | "(3772, 6319)\n" | ||
| 86 | ] | ||
| 87 | } | ||
| 88 | ], | ||
| 89 | "source": [ | ||
| 90 | "#NBI calculation for A\n", | ||
| 91 | "\n", | ||
| 92 | "Ky = np.diag((1/sum(A))) \n", | ||
| 93 | "n = A.shape[0]\n", | ||
| 94 | "m = A.shape[1]\n", | ||
| 95 | "#print n, m, Ky.shape\n", | ||
| 96 | "Ky[np.isinf(Ky) | np.isnan(Ky)] = 0\n", | ||
| 97 | "kx = np.transpose(np.sum(A,1))\n", | ||
| 98 | "#print kx.shape\n", | ||
| 99 | "Nx = np.matlib.repmat(1/kx,n,1)\n", | ||
| 100 | "Nx[np.isinf(Nx) | np.isnan(Nx)] = 0\n", | ||
| 101 | "#kx[np.isinf(kx) | np.isnan(kx)] = 0\n", | ||
| 102 | "W = np.transpose(np.dot(A, Ky))\n", | ||
| 103 | "W1 = np.dot(A, W)\n", | ||
| 104 | "W2 = np.multiply(Nx, W1)\n", | ||
| 105 | "print W2.shape\n", | ||
| 106 | "NBIscore = np.dot(W2, A)\n", | ||
| 107 | "print NBIscore.shape" | ||
| 108 | ] | ||
| 109 | }, | ||
| 110 | { | ||
| 111 | "cell_type": "code", | ||
| 112 | "execution_count": 4, | ||
| 113 | "metadata": { | ||
| 114 | "collapsed": true | ||
| 115 | }, | ||
| 116 | "outputs": [], | ||
| 117 | "source": [ | ||
| 118 | "#NBI calculation for A_T\n", | ||
| 119 | "\n", | ||
| 120 | "Ky = np.diag((1/sum(A_T)))\n", | ||
| 121 | "n = A_T.shape[0]\n", | ||
| 122 | "m = A_T.shape[1]\n", | ||
| 123 | "#print n, m, Ky.shape\n", | ||
| 124 | "Ky[np.isinf(Ky) | np.isnan(Ky)] = 0\n", | ||
| 125 | "kx = np.transpose(np.sum(A_T,1))\n", | ||
| 126 | "#print kx.shape\n", | ||
| 127 | "Nx = np.matlib.repmat(1/kx,n,1)\n", | ||
| 128 | "Nx[np.isinf(Nx) | np.isnan(Nx)] = 0\n", | ||
| 129 | "#kx[np.isinf(kx) | np.isnan(kx)] = 0\n", | ||
| 130 | "W = np.transpose(np.dot(A_T, Ky))\n", | ||
| 131 | "W1 = np.dot(A_T, W)\n", | ||
| 132 | "W2 = np.multiply(Nx, W1)\n", | ||
| 133 | "NBIscore_T = np.dot(W2, A_T)" | ||
| 134 | ] | ||
| 135 | }, | ||
| 136 | { | ||
| 137 | "cell_type": "code", | ||
| 138 | "execution_count": 5, | ||
| 139 | "metadata": { | ||
| 140 | "collapsed": false | ||
| 141 | }, | ||
| 142 | "outputs": [], | ||
| 143 | "source": [ | ||
| 144 | "##Normalizing NBI scores\n", | ||
| 145 | "NBIscore = np.true_divide(NBIscore, np.max(NBIscore, axis=0))\n", | ||
| 146 | "NBIscore_T = np.true_divide(NBIscore_T, np.max(NBIscore_T, axis=0))" | ||
| 147 | ] | ||
| 148 | }, | ||
| 149 | { | ||
| 150 | "cell_type": "code", | ||
| 151 | "execution_count": 7, | ||
| 152 | "metadata": { | ||
| 153 | "collapsed": false | ||
| 154 | }, | ||
| 155 | "outputs": [], | ||
| 156 | "source": [ | ||
| 157 | "nbi_idx = np.argsort(NBIscore, axis=0)\n", | ||
| 158 | "nbi_idx_T= np.argsort(NBIscore_T, axis=0)" | ||
| 159 | ] | ||
| 160 | }, | ||
| 161 | { | ||
| 162 | "cell_type": "code", | ||
| 163 | "execution_count": 8, | ||
| 164 | "metadata": { | ||
| 165 | "collapsed": false | ||
| 166 | }, | ||
| 167 | "outputs": [], | ||
| 168 | "source": [ | ||
| 169 | "wp = open('predicted_targets_for_all_drugs_using_percent_diff_0.20_.csv','w')\n", | ||
| 170 | "for d in range(nd):\n", | ||
| 171 | " idx1 = nbi_idx[:,d]\n", | ||
| 172 | " #idx1 = idx1[::-1]\n", | ||
| 173 | " idx2 = A[:,d]\n", | ||
| 174 | " idx3 = idx2[idx1]\n", | ||
| 175 | " idx4 = np.where(idx3 == 0)[0]\n", | ||
| 176 | " p_targets_idx = idx1[idx4[-n:]]\n", | ||
| 177 | " p_targets_idx = p_targets_idx[::-1]\n", | ||
| 178 | " p_targets = NBIscore[p_targets_idx,d]\n", | ||
| 179 | " if p_targets[0] == 0.0:\n", | ||
| 180 | " continue\n", | ||
| 181 | " else:\n", | ||
| 182 | " p_diff = np.diff(p_targets)\n", | ||
| 183 | " th = p_targets[0]*0.20\n", | ||
| 184 | " th_f = p_targets[0]-th\n", | ||
| 185 | " f_idx = p_targets_idx[p_targets > th_f]\n", | ||
| 186 | " f_scores = p_targets[p_targets > th_f]\n", | ||
| 187 | " f_targets = uni_targetid[f_idx]\n", | ||
| 188 | " #print p_targets[0], th, th_f, p_targets[p_targets > th_f], p_targets\n", | ||
| 189 | " for i,t in enumerate(f_targets):\n", | ||
| 190 | " wp.write(uni_drugid[d] + ',' + t + ',' + str(f_scores[i]) + '\\n')\n", | ||
| 191 | "\n", | ||
| 192 | "wp.close()" | ||
| 193 | ] | ||
| 194 | } | ||
| 195 | ], | ||
| 196 | "metadata": { | ||
| 197 | "kernelspec": { | ||
| 198 | "display_name": "Python 2", | ||
| 199 | "language": "python", | ||
| 200 | "name": "python2" | ||
| 201 | }, | ||
| 202 | "language_info": { | ||
| 203 | "codemirror_mode": { | ||
| 204 | "name": "ipython", | ||
| 205 | "version": 2 | ||
| 206 | }, | ||
| 207 | "file_extension": ".py", | ||
| 208 | "mimetype": "text/x-python", | ||
| 209 | "name": "python", | ||
| 210 | "nbconvert_exporter": "python", | ||
| 211 | "pygments_lexer": "ipython2", | ||
| 212 | "version": "2.7.6" | ||
| 213 | } | ||
| 214 | }, | ||
| 215 | "nbformat": 4, | ||
| 216 | "nbformat_minor": 2 | ||
| 217 | } | ||
| 218 |