465 lines (464 with data), 122.7 kB

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "os.chdir('../')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "RDKit WARNING: [17:12:07] Enabling RDKit 2019.09.3 jupyter extensions\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Beginning Processing...\n",
      "Loading customized repurposing dataset...\n",
      "Beginning Downloading Configs Files for training from scratch...\n",
      "Downloading finished... Beginning to extract zip file...\n",
      "Configs Models Successfully Downloaded...\n",
      "Training on your own customized data...\n",
      "in total: 26640 drug-target pairs\n",
      "encoding drug...\n",
      "unique drugs: 13763\n",
      "drug encoding finished...\n",
      "encoding protein...\n",
      "unique target sequence: 1\n",
      "protein encoding finished...\n",
      "splitting dataset...\n",
      "Done.\n",
      "Training from scrtach...\n",
      "Begin to train model 0 with drug encoding MPNN and target encoding CNN\n",
      "Let's use 1 GPU!\n",
      "--- Data Preparation ---\n",
      "--- Go for Training ---\n",
      "Training at Epoch 1 iteration 0 with loss 0.69219. Total time 0.00055 hours\n",
      "Training at Epoch 1 iteration 100 with loss 0.67406. Total time 0.01333 hours\n",
      "Validation at Epoch 1 , AUROC: 0.70764 , AUPRC: 0.08025 , F1: 0.09424\n",
      "Training at Epoch 2 iteration 0 with loss 0.62362. Total time 0.02305 hours\n",
      "Training at Epoch 2 iteration 100 with loss 0.52807. Total time 0.03611 hours\n",
      "Validation at Epoch 2 , AUROC: 0.72531 , AUPRC: 0.16038 , F1: 0.10389\n",
      "Training at Epoch 3 iteration 0 with loss 0.47976. Total time 0.04583 hours\n",
      "Training at Epoch 3 iteration 100 with loss 0.46717. Total time 0.05861 hours\n",
      "Validation at Epoch 3 , AUROC: 0.73128 , AUPRC: 0.16780 , F1: 0.13793\n",
      "Training at Epoch 4 iteration 0 with loss 0.47900. Total time 0.06833 hours\n",
      "Training at Epoch 4 iteration 100 with loss 0.40212. Total time 0.08138 hours\n",
      "Validation at Epoch 4 , AUROC: 0.73826 , AUPRC: 0.19690 , F1: 0.12826\n",
      "Training at Epoch 5 iteration 0 with loss 0.40222. Total time 0.09111 hours\n",
      "Training at Epoch 5 iteration 100 with loss 0.23715. Total time 0.10416 hours\n",
      "Validation at Epoch 5 , AUROC: 0.72835 , AUPRC: 0.23509 , F1: 0.23602\n",
      "Training at Epoch 6 iteration 0 with loss 0.27357. Total time 0.11388 hours\n",
      "Training at Epoch 6 iteration 100 with loss 0.35282. Total time 0.12666 hours\n",
      "Validation at Epoch 6 , AUROC: 0.72309 , AUPRC: 0.22652 , F1: 0.21229\n",
      "Training at Epoch 7 iteration 0 with loss 0.08800. Total time 0.13638 hours\n",
      "Training at Epoch 7 iteration 100 with loss 0.21136. Total time 0.14944 hours\n",
      "Validation at Epoch 7 , AUROC: 0.74493 , AUPRC: 0.19183 , F1: 0.25210\n",
      "Training at Epoch 8 iteration 0 with loss 0.17881. Total time 0.15916 hours\n",
      "Training at Epoch 8 iteration 100 with loss 0.11592. Total time 0.17194 hours\n",
      "Validation at Epoch 8 , AUROC: 0.72775 , AUPRC: 0.23099 , F1: 0.25742\n",
      "Training at Epoch 9 iteration 0 with loss 0.11575. Total time 0.18166 hours\n",
      "Training at Epoch 9 iteration 100 with loss 0.14362. Total time 0.19444 hours\n",
      "Validation at Epoch 9 , AUROC: 0.73358 , AUPRC: 0.21052 , F1: 0.23913\n",
      "Training at Epoch 10 iteration 0 with loss 0.17836. Total time 0.20416 hours\n",
      "Training at Epoch 10 iteration 100 with loss 0.08284. Total time 0.21722 hours\n",
      "Validation at Epoch 10 , AUROC: 0.72196 , AUPRC: 0.20493 , F1: 0.25490\n",
      "--- Go for Testing ---\n",
      "Testing AUROC: 0.794694459119799 , AUPRC: 0.29236629647158663 , F1: 0.30630630630630634\n",
      "--- Training Finished ---\n",
      "model training finished, now repurposing\n",
      "repurposing...\n",
      "in total: 82 drug-target pairs\n",
      "encoding drug...\n",
      "unique drugs: 81\n",
      "drug encoding finished...\n",
      "encoding protein...\n",
      "unique target sequence: 1\n",
      "protein encoding finished...\n",
      "Done.\n",
      "predicting...\n",
      "---------------\n",
      "Predictions from model 0 with drug encoding MPNN and target encoding CNN are done...\n",
      "in total: 26640 drug-target pairs\n",
      "encoding drug...\n",
      "unique drugs: 13763\n",
      "drug encoding finished...\n",
      "encoding protein...\n",
      "unique target sequence: 1\n",
      "protein encoding finished...\n",
      "splitting dataset...\n",
      "Done.\n",
      "Training from scrtach...\n",
      "Begin to train model 1 with drug encoding CNN and target encoding CNN\n",
      "Let's use 1 GPU!\n",
      "--- Data Preparation ---\n",
      "--- Go for Training ---\n",
      "Training at Epoch 1 iteration 0 with loss 0.69552. Total time 0.00027 hours\n",
      "Training at Epoch 1 iteration 100 with loss 0.60612. Total time 0.00833 hours\n",
      "Validation at Epoch 1 , AUROC: 0.72360 , AUPRC: 0.23331 , F1: 0.10069\n",
      "Training at Epoch 2 iteration 0 with loss 0.53683. Total time 0.01472 hours\n",
      "Training at Epoch 2 iteration 100 with loss 0.33064. Total time 0.02277 hours\n",
      "Validation at Epoch 2 , AUROC: 0.74836 , AUPRC: 0.22101 , F1: 0.19\n",
      "Training at Epoch 3 iteration 0 with loss 0.23112. Total time 0.02916 hours\n",
      "Training at Epoch 3 iteration 100 with loss 0.03467. Total time 0.0375 hours\n",
      "Validation at Epoch 3 , AUROC: 0.73631 , AUPRC: 0.24006 , F1: 0.25581\n",
      "Training at Epoch 4 iteration 0 with loss 0.09362. Total time 0.04388 hours\n",
      "Training at Epoch 4 iteration 100 with loss 0.02818. Total time 0.05194 hours\n",
      "Validation at Epoch 4 , AUROC: 0.75143 , AUPRC: 0.25542 , F1: 0.29545\n",
      "Training at Epoch 5 iteration 0 with loss 0.03108. Total time 0.05833 hours\n",
      "Training at Epoch 5 iteration 100 with loss 0.00439. Total time 0.06638 hours\n",
      "Validation at Epoch 5 , AUROC: 0.74331 , AUPRC: 0.26049 , F1: 0.19626\n",
      "Training at Epoch 6 iteration 0 with loss 0.09016. Total time 0.07277 hours\n",
      "Training at Epoch 6 iteration 100 with loss 0.04769. Total time 0.08083 hours\n",
      "Validation at Epoch 6 , AUROC: 0.73865 , AUPRC: 0.24654 , F1: 0.19999\n",
      "Training at Epoch 7 iteration 0 with loss 0.00803. Total time 0.08722 hours\n",
      "Training at Epoch 7 iteration 100 with loss 0.11943. Total time 0.09527 hours\n",
      "Validation at Epoch 7 , AUROC: 0.73311 , AUPRC: 0.22922 , F1: 0.26470\n",
      "Training at Epoch 8 iteration 0 with loss 0.01275. Total time 0.10166 hours\n",
      "Training at Epoch 8 iteration 100 with loss 0.01197. Total time 0.10972 hours\n",
      "Validation at Epoch 8 , AUROC: 0.75380 , AUPRC: 0.21910 , F1: 0.25\n",
      "Training at Epoch 9 iteration 0 with loss 0.00258. Total time 0.11611 hours\n",
      "Training at Epoch 9 iteration 100 with loss 0.02608. Total time 0.12416 hours\n",
      "Validation at Epoch 9 , AUROC: 0.74069 , AUPRC: 0.27212 , F1: 0.26262\n",
      "Training at Epoch 10 iteration 0 with loss 0.01071. Total time 0.13055 hours\n",
      "Training at Epoch 10 iteration 100 with loss 0.01437. Total time 0.13861 hours\n",
      "Validation at Epoch 10 , AUROC: 0.73440 , AUPRC: 0.25744 , F1: 0.27906\n",
      "--- Go for Testing ---\n",
      "Testing AUROC: 0.7888349903142661 , AUPRC: 0.32125016124670125 , F1: 0.37777777777777777\n",
      "--- Training Finished ---\n",
      "model training finished, now repurposing\n",
      "repurposing...\n",
      "in total: 82 drug-target pairs\n",
      "encoding drug...\n",
      "unique drugs: 81\n",
      "drug encoding finished...\n",
      "encoding protein...\n",
      "unique target sequence: 1\n",
      "protein encoding finished...\n",
      "Done.\n",
      "predicting...\n",
      "---------------\n",
      "Predictions from model 1 with drug encoding CNN and target encoding CNN are done...\n",
      "in total: 26640 drug-target pairs\n",
      "encoding drug...\n",
      "unique drugs: 13763\n",
      "drug encoding finished...\n",
      "encoding protein...\n",
      "unique target sequence: 1\n",
      "protein encoding finished...\n",
      "splitting dataset...\n",
      "Done.\n",
      "Training from scrtach...\n",
      "Begin to train model 2 with drug encoding Morgan and target encoding CNN\n",
      "Let's use 1 GPU!\n",
      "--- Data Preparation ---\n",
      "--- Go for Training ---\n",
      "Training at Epoch 1 iteration 0 with loss 0.69251. Total time 0.00027 hours\n",
      "Training at Epoch 1 iteration 100 with loss 0.03058. Total time 0.00666 hours\n",
      "Validation at Epoch 1 , AUROC: 0.70575 , AUPRC: 0.21219 , F1: 0.3125\n",
      "Training at Epoch 2 iteration 0 with loss 0.00571. Total time 0.01194 hours\n",
      "Training at Epoch 2 iteration 100 with loss 0.00601. Total time 0.01833 hours\n",
      "Validation at Epoch 2 , AUROC: 0.73703 , AUPRC: 0.27978 , F1: 0.26666\n",
      "Training at Epoch 3 iteration 0 with loss 0.00014. Total time 0.02361 hours\n",
      "Training at Epoch 3 iteration 100 with loss 0.00495. Total time 0.03 hours\n",
      "Validation at Epoch 3 , AUROC: 0.69870 , AUPRC: 0.16020 , F1: 0.21333\n",
      "Training at Epoch 4 iteration 0 with loss 0.12378. Total time 0.03527 hours\n",
      "Training at Epoch 4 iteration 100 with loss 0.00010. Total time 0.04166 hours\n",
      "Validation at Epoch 4 , AUROC: 0.73478 , AUPRC: 0.19567 , F1: 0.2\n",
      "Training at Epoch 5 iteration 0 with loss 0.00012. Total time 0.04666 hours\n",
      "Training at Epoch 5 iteration 100 with loss 0.00428. Total time 0.05333 hours\n",
      "Validation at Epoch 5 , AUROC: 0.71250 , AUPRC: 0.23266 , F1: 0.36585\n",
      "Training at Epoch 6 iteration 0 with loss 0.00166. Total time 0.05861 hours\n",
      "Training at Epoch 6 iteration 100 with loss 0.05938. Total time 0.06527 hours\n",
      "Validation at Epoch 6 , AUROC: 0.72070 , AUPRC: 0.25071 , F1: 0.29508\n",
      "Training at Epoch 7 iteration 0 with loss 8.21252. Total time 0.07027 hours\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training at Epoch 7 iteration 100 with loss 0.00050. Total time 0.07694 hours\n",
      "Validation at Epoch 7 , AUROC: 0.74203 , AUPRC: 0.25694 , F1: 0.32876\n",
      "Training at Epoch 8 iteration 0 with loss 0.00028. Total time 0.08194 hours\n",
      "Training at Epoch 8 iteration 100 with loss 8.92719. Total time 0.08861 hours\n",
      "Validation at Epoch 8 , AUROC: 0.76748 , AUPRC: 0.26690 , F1: 0.30769\n",
      "Training at Epoch 9 iteration 0 with loss 0.00865. Total time 0.09361 hours\n",
      "Training at Epoch 9 iteration 100 with loss 3.01948. Total time 0.1 hours\n",
      "Validation at Epoch 9 , AUROC: 0.75496 , AUPRC: 0.26952 , F1: 0.27118\n",
      "Training at Epoch 10 iteration 0 with loss 4.84512. Total time 0.10527 hours\n",
      "Training at Epoch 10 iteration 100 with loss 1.63913. Total time 0.11166 hours\n",
      "Validation at Epoch 10 , AUROC: 0.73905 , AUPRC: 0.20806 , F1: 0.30769\n",
      "--- Go for Testing ---\n",
      "Testing AUROC: 0.786321502329379 , AUPRC: 0.33377574030682805 , F1: 0.3661971830985916\n",
      "--- Training Finished ---\n",
      "model training finished, now repurposing\n",
      "repurposing...\n",
      "in total: 82 drug-target pairs\n",
      "encoding drug...\n",
      "unique drugs: 81\n",
      "drug encoding finished...\n",
      "encoding protein...\n",
      "unique target sequence: 1\n",
      "protein encoding finished...\n",
      "Done.\n",
      "predicting...\n",
      "---------------\n",
      "Predictions from model 2 with drug encoding Morgan and target encoding CNN are done...\n",
      "in total: 26640 drug-target pairs\n",
      "encoding drug...\n",
      "unique drugs: 13763\n",
      "drug encoding finished...\n",
      "encoding protein...\n",
      "unique target sequence: 1\n",
      "-- Encoding AAC takes time. Time Reference: 24s for ~100 sequences in a CPU. Calculate your time by the unique target sequence #, instead of the entire dataset.\n",
      "protein encoding finished...\n",
      "splitting dataset...\n",
      "Done.\n",
      "Training from scrtach...\n",
      "Begin to train model 3 with drug encoding Morgan and target encoding AAC\n",
      "Let's use 1 GPU!\n",
      "--- Data Preparation ---\n",
      "--- Go for Training ---\n",
      "Training at Epoch 1 iteration 0 with loss 0.69258. Total time 0.0 hours\n",
      "Training at Epoch 1 iteration 100 with loss 0.03368. Total time 0.00222 hours\n",
      "Validation at Epoch 1 , AUROC: 0.72777 , AUPRC: 0.23005 , F1: 0.27848\n",
      "Training at Epoch 2 iteration 0 with loss 0.05851. Total time 0.00416 hours\n",
      "Training at Epoch 2 iteration 100 with loss 0.00342. Total time 0.00638 hours\n",
      "Validation at Epoch 2 , AUROC: 0.72116 , AUPRC: 0.19938 , F1: 0.23188\n",
      "Training at Epoch 3 iteration 0 with loss 0.03552. Total time 0.00833 hours\n",
      "Training at Epoch 3 iteration 100 with loss 0.03714. Total time 0.01055 hours\n",
      "Validation at Epoch 3 , AUROC: 0.75044 , AUPRC: 0.24382 , F1: 0.26666\n",
      "Training at Epoch 4 iteration 0 with loss 0.00824. Total time 0.0125 hours\n",
      "Training at Epoch 4 iteration 100 with loss 2.96523. Total time 0.015 hours\n",
      "Validation at Epoch 4 , AUROC: 0.74432 , AUPRC: 0.22218 , F1: 0.22580\n",
      "Training at Epoch 5 iteration 0 with loss 0.00015. Total time 0.01694 hours\n",
      "Training at Epoch 5 iteration 100 with loss 4.28601. Total time 0.01916 hours\n",
      "Validation at Epoch 5 , AUROC: 0.73850 , AUPRC: 0.22437 , F1: 0.27692\n",
      "Training at Epoch 6 iteration 0 with loss 0.00086. Total time 0.02111 hours\n",
      "Training at Epoch 6 iteration 100 with loss 0.00075. Total time 0.02333 hours\n",
      "Validation at Epoch 6 , AUROC: 0.72299 , AUPRC: 0.16720 , F1: 0.11111\n",
      "Training at Epoch 7 iteration 0 with loss 0.00180. Total time 0.02527 hours\n",
      "Training at Epoch 7 iteration 100 with loss 0.01561. Total time 0.0275 hours\n",
      "Validation at Epoch 7 , AUROC: 0.68612 , AUPRC: 0.21516 , F1: 0.28571\n",
      "Training at Epoch 8 iteration 0 with loss 0.00058. Total time 0.02944 hours\n",
      "Training at Epoch 8 iteration 100 with loss 1.33094. Total time 0.03194 hours\n",
      "Validation at Epoch 8 , AUROC: 0.72089 , AUPRC: 0.21749 , F1: 0.27118\n",
      "Training at Epoch 9 iteration 0 with loss 7.78595. Total time 0.03361 hours\n",
      "Training at Epoch 9 iteration 100 with loss 0.01024. Total time 0.03611 hours\n",
      "Validation at Epoch 9 , AUROC: 0.70943 , AUPRC: 0.21453 , F1: 0.24778\n",
      "Training at Epoch 10 iteration 0 with loss 0.00707. Total time 0.03805 hours\n",
      "Training at Epoch 10 iteration 100 with loss 0.07515. Total time 0.04027 hours\n",
      "Validation at Epoch 10 , AUROC: 0.70842 , AUPRC: 0.17618 , F1: 0.22580\n",
      "--- Go for Testing ---\n",
      "Testing AUROC: 0.8021628804252117 , AUPRC: 0.2416151102959611 , F1: 0.2727272727272727\n",
      "--- Training Finished ---\n",
      "model training finished, now repurposing\n",
      "repurposing...\n",
      "in total: 82 drug-target pairs\n",
      "encoding drug...\n",
      "unique drugs: 81\n",
      "drug encoding finished...\n",
      "encoding protein...\n",
      "unique target sequence: 1\n",
      "-- Encoding AAC takes time. Time Reference: 24s for ~100 sequences in a CPU. Calculate your time by the unique target sequence #, instead of the entire dataset.\n",
      "protein encoding finished...\n",
      "Done.\n",
      "predicting...\n",
      "---------------\n",
      "Predictions from model 3 with drug encoding Morgan and target encoding AAC are done...\n",
      "in total: 26640 drug-target pairs\n",
      "encoding drug...\n",
      "unique drugs: 13763\n",
      "drug encoding finished...\n",
      "encoding protein...\n",
      "unique target sequence: 1\n",
      "-- Encoding AAC takes time. Time Reference: 24s for ~100 sequences in a CPU. Calculate your time by the unique target sequence #, instead of the entire dataset.\n",
      "protein encoding finished...\n",
      "splitting dataset...\n",
      "Done.\n",
      "Training from scrtach...\n",
      "Begin to train model 4 with drug encoding Daylight and target encoding AAC\n",
      "Let's use 1 GPU!\n",
      "--- Data Preparation ---\n",
      "--- Go for Training ---\n",
      "Training at Epoch 1 iteration 0 with loss 0.69373. Total time 0.0 hours\n",
      "Training at Epoch 1 iteration 100 with loss 0.24585. Total time 0.0025 hours\n",
      "Validation at Epoch 1 , AUROC: 0.72440 , AUPRC: 0.25495 , F1: 0.27722\n",
      "Training at Epoch 2 iteration 0 with loss 0.09313. Total time 0.00444 hours\n",
      "Training at Epoch 2 iteration 100 with loss 0.01475. Total time 0.00666 hours\n",
      "Validation at Epoch 2 , AUROC: 0.72943 , AUPRC: 0.24917 , F1: 0.26966\n",
      "Training at Epoch 3 iteration 0 with loss 0.01184. Total time 0.00861 hours\n",
      "Training at Epoch 3 iteration 100 with loss 0.04135. Total time 0.01111 hours\n",
      "Validation at Epoch 3 , AUROC: 0.73303 , AUPRC: 0.27552 , F1: 0.31578\n",
      "Training at Epoch 4 iteration 0 with loss 0.00597. Total time 0.01305 hours\n",
      "Training at Epoch 4 iteration 100 with loss 0.09914. Total time 0.01527 hours\n",
      "Validation at Epoch 4 , AUROC: 0.69888 , AUPRC: 0.25823 , F1: 0.30769\n",
      "Training at Epoch 5 iteration 0 with loss 0.01548. Total time 0.01722 hours\n",
      "Training at Epoch 5 iteration 100 with loss 0.03572. Total time 0.01972 hours\n",
      "Validation at Epoch 5 , AUROC: 0.72719 , AUPRC: 0.22114 , F1: 0.22222\n",
      "Training at Epoch 6 iteration 0 with loss 0.00811. Total time 0.02166 hours\n",
      "Training at Epoch 6 iteration 100 with loss 0.03094. Total time 0.02388 hours\n",
      "Validation at Epoch 6 , AUROC: 0.69107 , AUPRC: 0.20544 , F1: 0.24528\n",
      "Training at Epoch 7 iteration 0 with loss 0.11622. Total time 0.02611 hours\n",
      "Training at Epoch 7 iteration 100 with loss 0.00033. Total time 0.02833 hours\n",
      "Validation at Epoch 7 , AUROC: 0.68711 , AUPRC: 0.21855 , F1: 0.27160\n",
      "Training at Epoch 8 iteration 0 with loss 0.00479. Total time 0.03027 hours\n",
      "Training at Epoch 8 iteration 100 with loss 0.00167. Total time 0.03277 hours\n",
      "Validation at Epoch 8 , AUROC: 0.69214 , AUPRC: 0.18400 , F1: 0.29213\n",
      "Training at Epoch 9 iteration 0 with loss 0.01625. Total time 0.03472 hours\n",
      "Training at Epoch 9 iteration 100 with loss 0.02266. Total time 0.03694 hours\n",
      "Validation at Epoch 9 , AUROC: 0.71376 , AUPRC: 0.26114 , F1: 0.27659\n",
      "Training at Epoch 10 iteration 0 with loss 0.01236. Total time 0.03888 hours\n",
      "Training at Epoch 10 iteration 100 with loss 0.00125. Total time 0.04138 hours\n",
      "Validation at Epoch 10 , AUROC: 0.72438 , AUPRC: 0.23710 , F1: 0.36363\n",
      "--- Go for Testing ---\n",
      "Testing AUROC: 0.7898595968813537 , AUPRC: 0.4240506983064935 , F1: 0.43478260869565216\n",
      "--- Training Finished ---\n",
      "model training finished, now repurposing\n",
      "repurposing...\n",
      "in total: 82 drug-target pairs\n",
      "encoding drug...\n",
      "unique drugs: 81\n",
      "drug encoding finished...\n",
      "encoding protein...\n",
      "unique target sequence: 1\n",
      "-- Encoding AAC takes time. Time Reference: 24s for ~100 sequences in a CPU. Calculate your time by the unique target sequence #, instead of the entire dataset.\n",
      "protein encoding finished...\n",
      "Done.\n",
      "predicting...\n",
      "---------------\n",
      "Predictions from model 4 with drug encoding Daylight and target encoding AAC are done...\n",
      "models prediction finished...\n",
      "aggregating results...\n",
      "---------------\n",
      "Drug Repurposing Result for SARS-CoV 3CL Protease\n",
      "+------+----------------------+-----------------------+-------------+-------------+\n",
      "| Rank |      Drug Name       |      Target Name      | Interaction | Probability |\n",
      "+------+----------------------+-----------------------+-------------+-------------+\n",
      "|  1   |      Efavirenz       | SARS-CoV 3CL Protease |     YES     |     0.57    |\n",
      "|  2   |      Remdesivir      | SARS-CoV 3CL Protease |      NO     |     0.23    |\n",
      "|  3   |      Zanamivir       | SARS-CoV 3CL Protease |      NO     |     0.20    |\n",
      "|  4   |      Letermovir      | SARS-CoV 3CL Protease |      NO     |     0.13    |\n",
      "|  5   |   Podophyllotoxin    | SARS-CoV 3CL Protease |      NO     |     0.11    |\n",
      "|  6   |     Methisazone      | SARS-CoV 3CL Protease |      NO     |     0.06    |\n",
      "|  7   |      Tipranavir      | SARS-CoV 3CL Protease |      NO     |     0.02    |\n",
      "|  8   |      Atazanavir      | SARS-CoV 3CL Protease |      NO     |     0.01    |\n",
      "|  9   |     Elvitegravir     | SARS-CoV 3CL Protease |      NO     |     0.01    |\n",
      "|  10  |       Loviride       | SARS-CoV 3CL Protease |      NO     |     0.01    |\n",
      "|  11  |      Baloxavir       | SARS-CoV 3CL Protease |      NO     |     0.01    |\n",
      "|  12  |     Enfuvirtide      | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  13  |     Nitazoxanide     | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  14  |      Indinavir       | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  15  |      Darunavir       | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  16  |     Dolutegravir     | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  17  |      Amprenavir      | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  18  |      Pyrimidine      | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  19  |      Doravirine      | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  20  |      Vicriviroc      | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  21  |      Foscarnet       | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  22  |    Fosamprenavir     | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  23  |     Delavirdine      | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  24  |      Pleconaril      | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  25  |       Arbidol        | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  26  |     Taribavirin      | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  27  | Tenofovir_disoproxil | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  28  |     Bictegravir      | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  29  |     Rilpivirine      | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "|  30  |      Cidofovir       | SARS-CoV 3CL Protease |      NO     |     0.00    |\n",
      "checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from DeepPurpose import oneliner\n",
    "from DeepPurpose.dataset import *\n",
    "\n",
    "FILE_PATH = load_AID1706_txt_file()\n",
    "\n",
    "oneliner.repurpose(*load_SARS_CoV_Protease_3CL(), *load_antiviral_drugs(no_cid = True),  *read_file_training_dataset_bioassay(FILE_PATH), \\\n",
    "    split='HTS', convert_y = False, frac=[0.8,0.1,0.1], finetune_batch_size = 128, pretrained = False, agg = 'agg_mean_max')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}