Datasets
Models
Downloads: 1
2470 lines (2469 with data), 121.0 kB
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import os\n",
"os.chdir('../')"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"import DeepPurpose.oneliner as oneliner\n",
"from DeepPurpose import dataset\n",
"import time"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"X_repurpose, drug_names, drug_CID = dataset.load_antiviral_drugs('./data')"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for SARS-CoV2 3CL Protease\n",
"+------+----------------------+------------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+------------------------+---------------+\n",
"| 1 | Sofosbuvir | SARS-CoV2 3CL Protease | 360.22 |\n",
"| 2 | Daclatasvir | SARS-CoV2 3CL Protease | 424.06 |\n",
"| 3 | Vicriviroc | SARS-CoV2 3CL Protease | 623.78 |\n",
"| 4 | Efavirenz | SARS-CoV2 3CL Protease | 768.33 |\n",
"| 5 | Simeprevir | SARS-CoV2 3CL Protease | 781.29 |\n",
"| 6 | Etravirine | SARS-CoV2 3CL Protease | 809.88 |\n",
"| 7 | Amantadine | SARS-CoV2 3CL Protease | 826.28 |\n",
"| 8 | Letermovir | SARS-CoV2 3CL Protease | 891.66 |\n",
"| 9 | Rilpivirine | SARS-CoV2 3CL Protease | 929.63 |\n",
"| 10 | Ritonavir | SARS-CoV2 3CL Protease | 941.13 |\n",
"| 11 | Darunavir | SARS-CoV2 3CL Protease | 944.10 |\n",
"| 12 | Maraviroc | SARS-CoV2 3CL Protease | 945.22 |\n",
"| 13 | Lopinavir | SARS-CoV2 3CL Protease | 945.71 |\n",
"| 14 | Fosamprenavir | SARS-CoV2 3CL Protease | 964.99 |\n",
"| 15 | Peramivir | SARS-CoV2 3CL Protease | 1050.64 |\n",
"| 16 | Grazoprevir | SARS-CoV2 3CL Protease | 1202.52 |\n",
"| 17 | Amprenavir | SARS-CoV2 3CL Protease | 1204.21 |\n",
"| 18 | Telaprevir | SARS-CoV2 3CL Protease | 1212.97 |\n",
"| 19 | Elvitegravir | SARS-CoV2 3CL Protease | 1220.48 |\n",
"| 20 | Tenofovir | SARS-CoV2 3CL Protease | 1250.70 |\n",
"| 21 | Descovy | SARS-CoV2 3CL Protease | 1250.70 |\n",
"| 22 | Atazanavir | SARS-CoV2 3CL Protease | 1348.65 |\n",
"| 23 | Tromantadine | SARS-CoV2 3CL Protease | 1380.71 |\n",
"| 24 | Nelfinavir | SARS-CoV2 3CL Protease | 1451.43 |\n",
"| 25 | Abacavir | SARS-CoV2 3CL Protease | 1464.89 |\n",
"| 26 | Tenofovir_disoproxil | SARS-CoV2 3CL Protease | 1571.85 |\n",
"| 27 | Dolutegravir | SARS-CoV2 3CL Protease | 1672.82 |\n",
"| 28 | Delavirdine | SARS-CoV2 3CL Protease | 1691.47 |\n",
"| 29 | Saquinavir | SARS-CoV2 3CL Protease | 1763.63 |\n",
"| 30 | Raltegravir | SARS-CoV2 3CL Protease | 1854.40 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n",
"Time lapse:6.697427749633789\n"
]
}
],
"source": [
"start = time.time()\n",
"target, target_name = dataset.load_SARS_CoV2_Protease_3CL()\n",
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'mean')\n",
"end = time.time()\n",
"print('Time lapse:' + str(end - start))"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for SARS-CoV2 3CL Protease\n",
"+------+----------------------+------------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+------------------------+---------------+\n",
"| 1 | Lopinavir | SARS-CoV2 3CL Protease | 0.30 |\n",
"| 2 | Darunavir | SARS-CoV2 3CL Protease | 0.37 |\n",
"| 3 | Amprenavir | SARS-CoV2 3CL Protease | 1.31 |\n",
"| 4 | Tipranavir | SARS-CoV2 3CL Protease | 1.35 |\n",
"| 5 | Baloxavir | SARS-CoV2 3CL Protease | 1.69 |\n",
"| 6 | Boceprevir | SARS-CoV2 3CL Protease | 2.06 |\n",
"| 7 | Glecaprevir | SARS-CoV2 3CL Protease | 2.22 |\n",
"| 8 | Oseltamivir | SARS-CoV2 3CL Protease | 2.56 |\n",
"| 9 | Telaprevir | SARS-CoV2 3CL Protease | 2.70 |\n",
"| 10 | Nelfinavir | SARS-CoV2 3CL Protease | 3.56 |\n",
"| 11 | Maraviroc | SARS-CoV2 3CL Protease | 4.50 |\n",
"| 12 | Daclatasvir | SARS-CoV2 3CL Protease | 5.09 |\n",
"| 13 | Vicriviroc | SARS-CoV2 3CL Protease | 7.62 |\n",
"| 14 | Etravirine | SARS-CoV2 3CL Protease | 8.80 |\n",
"| 15 | Fosamprenavir | SARS-CoV2 3CL Protease | 9.91 |\n",
"| 16 | Entecavir | SARS-CoV2 3CL Protease | 10.25 |\n",
"| 17 | Atazanavir | SARS-CoV2 3CL Protease | 10.41 |\n",
"| 18 | Foscarnet | SARS-CoV2 3CL Protease | 11.34 |\n",
"| 19 | Simeprevir | SARS-CoV2 3CL Protease | 11.76 |\n",
"| 20 | Rilpivirine | SARS-CoV2 3CL Protease | 11.95 |\n",
"| 21 | Abacavir | SARS-CoV2 3CL Protease | 12.70 |\n",
"| 22 | Amantadine | SARS-CoV2 3CL Protease | 13.24 |\n",
"| 23 | Pleconaril | SARS-CoV2 3CL Protease | 13.74 |\n",
"| 24 | Saquinavir | SARS-CoV2 3CL Protease | 19.87 |\n",
"| 25 | Sofosbuvir | SARS-CoV2 3CL Protease | 20.28 |\n",
"| 26 | Delavirdine | SARS-CoV2 3CL Protease | 20.65 |\n",
"| 27 | Raltegravir | SARS-CoV2 3CL Protease | 22.45 |\n",
"| 28 | Tenofovir | SARS-CoV2 3CL Protease | 25.22 |\n",
"| 29 | Descovy | SARS-CoV2 3CL Protease | 25.22 |\n",
"| 30 | Peramivir | SARS-CoV2 3CL Protease | 25.57 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'max_effect')"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for SARS-CoV2 3CL Protease\n",
"+------+----------------------+------------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+------------------------+---------------+\n",
"| 1 | Sofosbuvir | SARS-CoV2 3CL Protease | 190.25 |\n",
"| 2 | Daclatasvir | SARS-CoV2 3CL Protease | 214.58 |\n",
"| 3 | Vicriviroc | SARS-CoV2 3CL Protease | 315.70 |\n",
"| 4 | Simeprevir | SARS-CoV2 3CL Protease | 396.53 |\n",
"| 5 | Etravirine | SARS-CoV2 3CL Protease | 409.34 |\n",
"| 6 | Amantadine | SARS-CoV2 3CL Protease | 419.76 |\n",
"| 7 | Letermovir | SARS-CoV2 3CL Protease | 460.28 |\n",
"| 8 | Rilpivirine | SARS-CoV2 3CL Protease | 470.79 |\n",
"| 9 | Darunavir | SARS-CoV2 3CL Protease | 472.24 |\n",
"| 10 | Lopinavir | SARS-CoV2 3CL Protease | 473.01 |\n",
"| 11 | Maraviroc | SARS-CoV2 3CL Protease | 474.86 |\n",
"| 12 | Fosamprenavir | SARS-CoV2 3CL Protease | 487.45 |\n",
"| 13 | Ritonavir | SARS-CoV2 3CL Protease | 492.19 |\n",
"| 14 | Efavirenz | SARS-CoV2 3CL Protease | 513.81 |\n",
"| 15 | Peramivir | SARS-CoV2 3CL Protease | 538.11 |\n",
"| 16 | Amprenavir | SARS-CoV2 3CL Protease | 602.76 |\n",
"| 17 | Telaprevir | SARS-CoV2 3CL Protease | 607.84 |\n",
"| 18 | Grazoprevir | SARS-CoV2 3CL Protease | 632.54 |\n",
"| 19 | Tenofovir | SARS-CoV2 3CL Protease | 637.96 |\n",
"| 20 | Descovy | SARS-CoV2 3CL Protease | 637.96 |\n",
"| 21 | Elvitegravir | SARS-CoV2 3CL Protease | 654.94 |\n",
"| 22 | Atazanavir | SARS-CoV2 3CL Protease | 679.53 |\n",
"| 23 | Nelfinavir | SARS-CoV2 3CL Protease | 727.49 |\n",
"| 24 | Abacavir | SARS-CoV2 3CL Protease | 738.80 |\n",
"| 25 | Tenofovir_disoproxil | SARS-CoV2 3CL Protease | 828.19 |\n",
"| 26 | Delavirdine | SARS-CoV2 3CL Protease | 856.06 |\n",
"| 27 | Tromantadine | SARS-CoV2 3CL Protease | 863.40 |\n",
"| 28 | Saquinavir | SARS-CoV2 3CL Protease | 891.75 |\n",
"| 29 | Dolutegravir | SARS-CoV2 3CL Protease | 920.32 |\n",
"| 30 | Raltegravir | SARS-CoV2 3CL Protease | 938.42 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'agg_mean_max')"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for RNA_polymerase_SARS_CoV2\n",
"+------+----------------------+--------------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+--------------------------+---------------+\n",
"| 1 | Daclatasvir | RNA_polymerase_SARS_CoV2 | 380.69 |\n",
"| 2 | Vicriviroc | RNA_polymerase_SARS_CoV2 | 404.62 |\n",
"| 3 | Simeprevir | RNA_polymerase_SARS_CoV2 | 408.03 |\n",
"| 4 | Sofosbuvir | RNA_polymerase_SARS_CoV2 | 647.43 |\n",
"| 5 | Etravirine | RNA_polymerase_SARS_CoV2 | 735.35 |\n",
"| 6 | Atazanavir | RNA_polymerase_SARS_CoV2 | 770.73 |\n",
"| 7 | Rilpivirine | RNA_polymerase_SARS_CoV2 | 906.63 |\n",
"| 8 | Maraviroc | RNA_polymerase_SARS_CoV2 | 911.97 |\n",
"| 9 | Letermovir | RNA_polymerase_SARS_CoV2 | 932.10 |\n",
"| 10 | Lopinavir | RNA_polymerase_SARS_CoV2 | 936.25 |\n",
"| 11 | Darunavir | RNA_polymerase_SARS_CoV2 | 936.39 |\n",
"| 12 | Fosamprenavir | RNA_polymerase_SARS_CoV2 | 945.02 |\n",
"| 13 | Peramivir | RNA_polymerase_SARS_CoV2 | 964.85 |\n",
"| 14 | Telaprevir | RNA_polymerase_SARS_CoV2 | 1123.34 |\n",
"| 15 | Amprenavir | RNA_polymerase_SARS_CoV2 | 1202.55 |\n",
"| 16 | Grazoprevir | RNA_polymerase_SARS_CoV2 | 1236.20 |\n",
"| 17 | Nelfinavir | RNA_polymerase_SARS_CoV2 | 1252.00 |\n",
"| 18 | Boceprevir | RNA_polymerase_SARS_CoV2 | 1391.99 |\n",
"| 19 | Raltegravir | RNA_polymerase_SARS_CoV2 | 1552.78 |\n",
"| 20 | Abacavir | RNA_polymerase_SARS_CoV2 | 1660.44 |\n",
"| 21 | Dolutegravir | RNA_polymerase_SARS_CoV2 | 1718.55 |\n",
"| 22 | Delavirdine | RNA_polymerase_SARS_CoV2 | 1746.95 |\n",
"| 23 | Doravirine | RNA_polymerase_SARS_CoV2 | 1763.03 |\n",
"| 24 | Elvitegravir | RNA_polymerase_SARS_CoV2 | 1821.02 |\n",
"| 25 | Saquinavir | RNA_polymerase_SARS_CoV2 | 1829.28 |\n",
"| 26 | Enfuvirtide | RNA_polymerase_SARS_CoV2 | 2177.03 |\n",
"| 27 | Pleconaril | RNA_polymerase_SARS_CoV2 | 2266.15 |\n",
"| 28 | Glecaprevir | RNA_polymerase_SARS_CoV2 | 2306.74 |\n",
"| 29 | Amantadine | RNA_polymerase_SARS_CoV2 | 2434.83 |\n",
"| 30 | Efavirenz | RNA_polymerase_SARS_CoV2 | 2617.99 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"target, target_name = dataset.load_SARS_CoV2_RNA_polymerase()\n",
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'mean')"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for RNA_polymerase_SARS_CoV2\n",
"+------+----------------------+--------------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+--------------------------+---------------+\n",
"| 1 | Lopinavir | RNA_polymerase_SARS_CoV2 | 0.28 |\n",
"| 2 | Darunavir | RNA_polymerase_SARS_CoV2 | 0.36 |\n",
"| 3 | Amprenavir | RNA_polymerase_SARS_CoV2 | 1.22 |\n",
"| 4 | Tipranavir | RNA_polymerase_SARS_CoV2 | 5.41 |\n",
"| 5 | Sofosbuvir | RNA_polymerase_SARS_CoV2 | 6.37 |\n",
"| 6 | Daclatasvir | RNA_polymerase_SARS_CoV2 | 6.66 |\n",
"| 7 | Baloxavir | RNA_polymerase_SARS_CoV2 | 7.39 |\n",
"| 8 | Pleconaril | RNA_polymerase_SARS_CoV2 | 7.54 |\n",
"| 9 | Boceprevir | RNA_polymerase_SARS_CoV2 | 8.10 |\n",
"| 10 | Vicriviroc | RNA_polymerase_SARS_CoV2 | 8.32 |\n",
"| 11 | Fosamprenavir | RNA_polymerase_SARS_CoV2 | 8.46 |\n",
"| 12 | Tenofovir | RNA_polymerase_SARS_CoV2 | 10.18 |\n",
"| 13 | Descovy | RNA_polymerase_SARS_CoV2 | 10.18 |\n",
"| 14 | Foscarnet | RNA_polymerase_SARS_CoV2 | 10.84 |\n",
"| 15 | Nelfinavir | RNA_polymerase_SARS_CoV2 | 11.13 |\n",
"| 16 | Oseltamivir | RNA_polymerase_SARS_CoV2 | 11.73 |\n",
"| 17 | Maraviroc | RNA_polymerase_SARS_CoV2 | 11.84 |\n",
"| 18 | Glecaprevir | RNA_polymerase_SARS_CoV2 | 11.87 |\n",
"| 19 | Amantadine | RNA_polymerase_SARS_CoV2 | 12.28 |\n",
"| 20 | Telaprevir | RNA_polymerase_SARS_CoV2 | 12.56 |\n",
"| 21 | Arbidol | RNA_polymerase_SARS_CoV2 | 14.80 |\n",
"| 22 | Remdesivir | RNA_polymerase_SARS_CoV2 | 18.93 |\n",
"| 23 | Letermovir | RNA_polymerase_SARS_CoV2 | 20.34 |\n",
"| 24 | Abacavir | RNA_polymerase_SARS_CoV2 | 24.28 |\n",
"| 25 | Saquinavir | RNA_polymerase_SARS_CoV2 | 25.48 |\n",
"| 26 | Rimantadine | RNA_polymerase_SARS_CoV2 | 37.71 |\n",
"| 27 | Rilpivirine | RNA_polymerase_SARS_CoV2 | 38.50 |\n",
"| 28 | Delavirdine | RNA_polymerase_SARS_CoV2 | 40.78 |\n",
"| 29 | Ritonavir | RNA_polymerase_SARS_CoV2 | 43.73 |\n",
"| 30 | Loviride | RNA_polymerase_SARS_CoV2 | 63.94 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'max_effect')"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for RNA_polymerase_SARS_CoV2\n",
"+------+----------------------+--------------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+--------------------------+---------------+\n",
"| 1 | Daclatasvir | RNA_polymerase_SARS_CoV2 | 193.68 |\n",
"| 2 | Vicriviroc | RNA_polymerase_SARS_CoV2 | 206.47 |\n",
"| 3 | Simeprevir | RNA_polymerase_SARS_CoV2 | 247.13 |\n",
"| 4 | Sofosbuvir | RNA_polymerase_SARS_CoV2 | 326.90 |\n",
"| 5 | Etravirine | RNA_polymerase_SARS_CoV2 | 420.95 |\n",
"| 6 | Atazanavir | RNA_polymerase_SARS_CoV2 | 422.32 |\n",
"| 7 | Maraviroc | RNA_polymerase_SARS_CoV2 | 461.91 |\n",
"| 8 | Lopinavir | RNA_polymerase_SARS_CoV2 | 468.27 |\n",
"| 9 | Darunavir | RNA_polymerase_SARS_CoV2 | 468.37 |\n",
"| 10 | Rilpivirine | RNA_polymerase_SARS_CoV2 | 472.57 |\n",
"| 11 | Letermovir | RNA_polymerase_SARS_CoV2 | 476.22 |\n",
"| 12 | Fosamprenavir | RNA_polymerase_SARS_CoV2 | 476.74 |\n",
"| 13 | Peramivir | RNA_polymerase_SARS_CoV2 | 515.97 |\n",
"| 14 | Telaprevir | RNA_polymerase_SARS_CoV2 | 567.95 |\n",
"| 15 | Amprenavir | RNA_polymerase_SARS_CoV2 | 601.88 |\n",
"| 16 | Nelfinavir | RNA_polymerase_SARS_CoV2 | 631.56 |\n",
"| 17 | Grazoprevir | RNA_polymerase_SARS_CoV2 | 657.71 |\n",
"| 18 | Boceprevir | RNA_polymerase_SARS_CoV2 | 700.05 |\n",
"| 19 | Abacavir | RNA_polymerase_SARS_CoV2 | 842.36 |\n",
"| 20 | Raltegravir | RNA_polymerase_SARS_CoV2 | 870.85 |\n",
"| 21 | Delavirdine | RNA_polymerase_SARS_CoV2 | 893.87 |\n",
"| 22 | Saquinavir | RNA_polymerase_SARS_CoV2 | 927.38 |\n",
"| 23 | Elvitegravir | RNA_polymerase_SARS_CoV2 | 983.21 |\n",
"| 24 | Doravirine | RNA_polymerase_SARS_CoV2 | 1034.73 |\n",
"| 25 | Dolutegravir | RNA_polymerase_SARS_CoV2 | 1096.64 |\n",
"| 26 | Pleconaril | RNA_polymerase_SARS_CoV2 | 1136.85 |\n",
"| 27 | Glecaprevir | RNA_polymerase_SARS_CoV2 | 1159.31 |\n",
"| 28 | Enfuvirtide | RNA_polymerase_SARS_CoV2 | 1212.25 |\n",
"| 29 | Amantadine | RNA_polymerase_SARS_CoV2 | 1223.55 |\n",
"| 30 | Ritonavir | RNA_polymerase_SARS_CoV2 | 1395.41 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'agg_mean_max')"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for SARS_CoV2_Helicase\n",
"+------+----------------------+--------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+--------------------+---------------+\n",
"| 1 | Daclatasvir | SARS_CoV2_Helicase | 422.51 |\n",
"| 2 | Simeprevir | SARS_CoV2_Helicase | 436.10 |\n",
"| 3 | Sofosbuvir | SARS_CoV2_Helicase | 460.44 |\n",
"| 4 | Vicriviroc | SARS_CoV2_Helicase | 624.43 |\n",
"| 5 | Etravirine | SARS_CoV2_Helicase | 749.41 |\n",
"| 6 | Atazanavir | SARS_CoV2_Helicase | 822.11 |\n",
"| 7 | Rilpivirine | SARS_CoV2_Helicase | 896.30 |\n",
"| 8 | Letermovir | SARS_CoV2_Helicase | 904.84 |\n",
"| 9 | Grazoprevir | SARS_CoV2_Helicase | 944.09 |\n",
"| 10 | Maraviroc | SARS_CoV2_Helicase | 958.09 |\n",
"| 11 | Lopinavir | SARS_CoV2_Helicase | 959.09 |\n",
"| 12 | Darunavir | SARS_CoV2_Helicase | 960.77 |\n",
"| 13 | Peramivir | SARS_CoV2_Helicase | 971.53 |\n",
"| 14 | Fosamprenavir | SARS_CoV2_Helicase | 982.04 |\n",
"| 15 | Tenofovir_disoproxil | SARS_CoV2_Helicase | 1025.72 |\n",
"| 16 | Amantadine | SARS_CoV2_Helicase | 1067.12 |\n",
"| 17 | Efavirenz | SARS_CoV2_Helicase | 1116.72 |\n",
"| 18 | Telaprevir | SARS_CoV2_Helicase | 1188.83 |\n",
"| 19 | Amprenavir | SARS_CoV2_Helicase | 1229.83 |\n",
"| 20 | Elvitegravir | SARS_CoV2_Helicase | 1338.24 |\n",
"| 21 | Nelfinavir | SARS_CoV2_Helicase | 1339.03 |\n",
"| 22 | Tenofovir | SARS_CoV2_Helicase | 1370.97 |\n",
"| 23 | Descovy | SARS_CoV2_Helicase | 1370.97 |\n",
"| 24 | Ritonavir | SARS_CoV2_Helicase | 1405.82 |\n",
"| 25 | Doravirine | SARS_CoV2_Helicase | 1477.56 |\n",
"| 26 | Abacavir | SARS_CoV2_Helicase | 1498.16 |\n",
"| 27 | Boceprevir | SARS_CoV2_Helicase | 1757.84 |\n",
"| 28 | Dolutegravir | SARS_CoV2_Helicase | 1764.69 |\n",
"| 29 | Pleconaril | SARS_CoV2_Helicase | 1787.40 |\n",
"| 30 | Delavirdine | SARS_CoV2_Helicase | 1796.55 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"target, target_name = dataset.load_SARS_CoV2_Helicase()\n",
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'mean')"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for SARS_CoV2_Helicase\n",
"+------+----------------------+--------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+--------------------+---------------+\n",
"| 1 | Lopinavir | SARS_CoV2_Helicase | 0.26 |\n",
"| 2 | Darunavir | SARS_CoV2_Helicase | 0.33 |\n",
"| 3 | Amprenavir | SARS_CoV2_Helicase | 0.96 |\n",
"| 4 | Tipranavir | SARS_CoV2_Helicase | 2.84 |\n",
"| 5 | Baloxavir | SARS_CoV2_Helicase | 4.24 |\n",
"| 6 | Boceprevir | SARS_CoV2_Helicase | 4.34 |\n",
"| 7 | Vicriviroc | SARS_CoV2_Helicase | 5.51 |\n",
"| 8 | Fosamprenavir | SARS_CoV2_Helicase | 5.58 |\n",
"| 9 | Oseltamivir | SARS_CoV2_Helicase | 5.73 |\n",
"| 10 | Glecaprevir | SARS_CoV2_Helicase | 5.78 |\n",
"| 11 | Telaprevir | SARS_CoV2_Helicase | 6.22 |\n",
"| 12 | Daclatasvir | SARS_CoV2_Helicase | 6.50 |\n",
"| 13 | Nelfinavir | SARS_CoV2_Helicase | 8.26 |\n",
"| 14 | Amantadine | SARS_CoV2_Helicase | 10.22 |\n",
"| 15 | Pleconaril | SARS_CoV2_Helicase | 11.02 |\n",
"| 16 | Maraviroc | SARS_CoV2_Helicase | 11.37 |\n",
"| 17 | Foscarnet | SARS_CoV2_Helicase | 11.44 |\n",
"| 18 | Sofosbuvir | SARS_CoV2_Helicase | 12.00 |\n",
"| 19 | Abacavir | SARS_CoV2_Helicase | 15.96 |\n",
"| 20 | Tenofovir | SARS_CoV2_Helicase | 18.68 |\n",
"| 21 | Descovy | SARS_CoV2_Helicase | 18.68 |\n",
"| 22 | Arbidol | SARS_CoV2_Helicase | 18.93 |\n",
"| 23 | Letermovir | SARS_CoV2_Helicase | 20.55 |\n",
"| 24 | Ritonavir | SARS_CoV2_Helicase | 26.00 |\n",
"| 25 | Rimantadine | SARS_CoV2_Helicase | 26.93 |\n",
"| 26 | Remdesivir | SARS_CoV2_Helicase | 27.36 |\n",
"| 27 | Atazanavir | SARS_CoV2_Helicase | 29.60 |\n",
"| 28 | Saquinavir | SARS_CoV2_Helicase | 29.99 |\n",
"| 29 | Simeprevir | SARS_CoV2_Helicase | 32.97 |\n",
"| 30 | Etravirine | SARS_CoV2_Helicase | 34.18 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'max_effect')"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for SARS_CoV2_Helicase\n",
"+------+----------------------+--------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+--------------------+---------------+\n",
"| 1 | Daclatasvir | SARS_CoV2_Helicase | 214.51 |\n",
"| 2 | Simeprevir | SARS_CoV2_Helicase | 234.54 |\n",
"| 3 | Sofosbuvir | SARS_CoV2_Helicase | 236.22 |\n",
"| 4 | Vicriviroc | SARS_CoV2_Helicase | 314.97 |\n",
"| 5 | Etravirine | SARS_CoV2_Helicase | 391.80 |\n",
"| 6 | Atazanavir | SARS_CoV2_Helicase | 425.85 |\n",
"| 7 | Letermovir | SARS_CoV2_Helicase | 462.70 |\n",
"| 8 | Rilpivirine | SARS_CoV2_Helicase | 474.99 |\n",
"| 9 | Lopinavir | SARS_CoV2_Helicase | 479.67 |\n",
"| 10 | Darunavir | SARS_CoV2_Helicase | 480.55 |\n",
"| 11 | Maraviroc | SARS_CoV2_Helicase | 484.73 |\n",
"| 12 | Fosamprenavir | SARS_CoV2_Helicase | 493.81 |\n",
"| 13 | Peramivir | SARS_CoV2_Helicase | 516.79 |\n",
"| 14 | Grazoprevir | SARS_CoV2_Helicase | 525.42 |\n",
"| 15 | Amantadine | SARS_CoV2_Helicase | 538.67 |\n",
"| 16 | Telaprevir | SARS_CoV2_Helicase | 597.52 |\n",
"| 17 | Amprenavir | SARS_CoV2_Helicase | 615.40 |\n",
"| 18 | Tenofovir_disoproxil | SARS_CoV2_Helicase | 620.33 |\n",
"| 19 | Nelfinavir | SARS_CoV2_Helicase | 673.65 |\n",
"| 20 | Efavirenz | SARS_CoV2_Helicase | 689.13 |\n",
"| 21 | Tenofovir | SARS_CoV2_Helicase | 694.82 |\n",
"| 22 | Descovy | SARS_CoV2_Helicase | 694.82 |\n",
"| 23 | Ritonavir | SARS_CoV2_Helicase | 715.91 |\n",
"| 24 | Elvitegravir | SARS_CoV2_Helicase | 729.73 |\n",
"| 25 | Abacavir | SARS_CoV2_Helicase | 757.06 |\n",
"| 26 | Doravirine | SARS_CoV2_Helicase | 800.64 |\n",
"| 27 | Boceprevir | SARS_CoV2_Helicase | 881.09 |\n",
"| 28 | Pleconaril | SARS_CoV2_Helicase | 899.21 |\n",
"| 29 | Delavirdine | SARS_CoV2_Helicase | 924.90 |\n",
"| 30 | Raltegravir | SARS_CoV2_Helicase | 934.50 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'agg_mean_max')"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for SARS_CoV2_3to5_exonuclease\n",
"+------+----------------------+----------------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+----------------------------+---------------+\n",
"| 1 | Sofosbuvir | SARS_CoV2_3to5_exonuclease | 331.42 |\n",
"| 2 | Simeprevir | SARS_CoV2_3to5_exonuclease | 357.64 |\n",
"| 3 | Daclatasvir | SARS_CoV2_3to5_exonuclease | 391.05 |\n",
"| 4 | Vicriviroc | SARS_CoV2_3to5_exonuclease | 511.83 |\n",
"| 5 | Atazanavir | SARS_CoV2_3to5_exonuclease | 669.99 |\n",
"| 6 | Etravirine | SARS_CoV2_3to5_exonuclease | 717.46 |\n",
"| 7 | Tenofovir_disoproxil | SARS_CoV2_3to5_exonuclease | 733.86 |\n",
"| 8 | Efavirenz | SARS_CoV2_3to5_exonuclease | 767.65 |\n",
"| 9 | Grazoprevir | SARS_CoV2_3to5_exonuclease | 820.81 |\n",
"| 10 | Rilpivirine | SARS_CoV2_3to5_exonuclease | 859.15 |\n",
"| 11 | Letermovir | SARS_CoV2_3to5_exonuclease | 865.38 |\n",
"| 12 | Peramivir | SARS_CoV2_3to5_exonuclease | 877.47 |\n",
"| 13 | Lopinavir | SARS_CoV2_3to5_exonuclease | 925.16 |\n",
"| 14 | Darunavir | SARS_CoV2_3to5_exonuclease | 930.19 |\n",
"| 15 | Maraviroc | SARS_CoV2_3to5_exonuclease | 933.72 |\n",
"| 16 | Fosamprenavir | SARS_CoV2_3to5_exonuclease | 952.93 |\n",
"| 17 | Elvitegravir | SARS_CoV2_3to5_exonuclease | 971.27 |\n",
"| 18 | Amantadine | SARS_CoV2_3to5_exonuclease | 977.99 |\n",
"| 19 | Telaprevir | SARS_CoV2_3to5_exonuclease | 1103.15 |\n",
"| 20 | Tenofovir | SARS_CoV2_3to5_exonuclease | 1111.73 |\n",
"| 21 | Descovy | SARS_CoV2_3to5_exonuclease | 1111.73 |\n",
"| 22 | Boceprevir | SARS_CoV2_3to5_exonuclease | 1137.67 |\n",
"| 23 | Nelfinavir | SARS_CoV2_3to5_exonuclease | 1189.33 |\n",
"| 24 | Amprenavir | SARS_CoV2_3to5_exonuclease | 1190.66 |\n",
"| 25 | Doravirine | SARS_CoV2_3to5_exonuclease | 1240.21 |\n",
"| 26 | Ritonavir | SARS_CoV2_3to5_exonuclease | 1310.21 |\n",
"| 27 | Abacavir | SARS_CoV2_3to5_exonuclease | 1424.62 |\n",
"| 28 | Raltegravir | SARS_CoV2_3to5_exonuclease | 1515.33 |\n",
"| 29 | Dolutegravir | SARS_CoV2_3to5_exonuclease | 1593.56 |\n",
"| 30 | Pleconaril | SARS_CoV2_3to5_exonuclease | 1645.66 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"target, target_name = dataset.load_SARS_CoV2_3to5_exonuclease()\n",
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'mean')"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for SARS_CoV2_3to5_exonuclease\n",
"+------+----------------------+----------------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+----------------------------+---------------+\n",
"| 1 | Lopinavir | SARS_CoV2_3to5_exonuclease | 0.21 |\n",
"| 2 | Darunavir | SARS_CoV2_3to5_exonuclease | 0.26 |\n",
"| 3 | Amprenavir | SARS_CoV2_3to5_exonuclease | 0.77 |\n",
"| 4 | Tipranavir | SARS_CoV2_3to5_exonuclease | 1.98 |\n",
"| 5 | Baloxavir | SARS_CoV2_3to5_exonuclease | 2.76 |\n",
"| 6 | Boceprevir | SARS_CoV2_3to5_exonuclease | 3.35 |\n",
"| 7 | Glecaprevir | SARS_CoV2_3to5_exonuclease | 3.63 |\n",
"| 8 | Oseltamivir | SARS_CoV2_3to5_exonuclease | 4.12 |\n",
"| 9 | Telaprevir | SARS_CoV2_3to5_exonuclease | 4.44 |\n",
"| 10 | Nelfinavir | SARS_CoV2_3to5_exonuclease | 5.15 |\n",
"| 11 | Daclatasvir | SARS_CoV2_3to5_exonuclease | 5.31 |\n",
"| 12 | Vicriviroc | SARS_CoV2_3to5_exonuclease | 5.57 |\n",
"| 13 | Fosamprenavir | SARS_CoV2_3to5_exonuclease | 5.64 |\n",
"| 14 | Maraviroc | SARS_CoV2_3to5_exonuclease | 7.09 |\n",
"| 15 | Amantadine | SARS_CoV2_3to5_exonuclease | 8.80 |\n",
"| 16 | Etravirine | SARS_CoV2_3to5_exonuclease | 10.17 |\n",
"| 17 | Foscarnet | SARS_CoV2_3to5_exonuclease | 11.20 |\n",
"| 18 | Entecavir | SARS_CoV2_3to5_exonuclease | 13.15 |\n",
"| 19 | Rilpivirine | SARS_CoV2_3to5_exonuclease | 14.35 |\n",
"| 20 | Atazanavir | SARS_CoV2_3to5_exonuclease | 14.42 |\n",
"| 21 | Simeprevir | SARS_CoV2_3to5_exonuclease | 14.67 |\n",
"| 22 | Sofosbuvir | SARS_CoV2_3to5_exonuclease | 15.18 |\n",
"| 23 | Pleconaril | SARS_CoV2_3to5_exonuclease | 15.20 |\n",
"| 24 | Abacavir | SARS_CoV2_3to5_exonuclease | 16.05 |\n",
"| 25 | Arbidol | SARS_CoV2_3to5_exonuclease | 18.36 |\n",
"| 26 | Saquinavir | SARS_CoV2_3to5_exonuclease | 19.92 |\n",
"| 27 | Tenofovir | SARS_CoV2_3to5_exonuclease | 20.45 |\n",
"| 28 | Descovy | SARS_CoV2_3to5_exonuclease | 20.45 |\n",
"| 29 | Ritonavir | SARS_CoV2_3to5_exonuclease | 26.42 |\n",
"| 30 | Letermovir | SARS_CoV2_3to5_exonuclease | 26.89 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'max_effect')"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for SARS_CoV2_3to5_exonuclease\n",
"+------+----------------------+----------------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+----------------------------+---------------+\n",
"| 1 | Sofosbuvir | SARS_CoV2_3to5_exonuclease | 173.30 |\n",
"| 2 | Simeprevir | SARS_CoV2_3to5_exonuclease | 186.16 |\n",
"| 3 | Daclatasvir | SARS_CoV2_3to5_exonuclease | 198.18 |\n",
"| 4 | Vicriviroc | SARS_CoV2_3to5_exonuclease | 258.70 |\n",
"| 5 | Atazanavir | SARS_CoV2_3to5_exonuclease | 342.21 |\n",
"| 6 | Etravirine | SARS_CoV2_3to5_exonuclease | 363.81 |\n",
"| 7 | Tenofovir_disoproxil | SARS_CoV2_3to5_exonuclease | 430.66 |\n",
"| 8 | Rilpivirine | SARS_CoV2_3to5_exonuclease | 436.75 |\n",
"| 9 | Letermovir | SARS_CoV2_3to5_exonuclease | 446.13 |\n",
"| 10 | Peramivir | SARS_CoV2_3to5_exonuclease | 456.39 |\n",
"| 11 | Lopinavir | SARS_CoV2_3to5_exonuclease | 462.68 |\n",
"| 12 | Grazoprevir | SARS_CoV2_3to5_exonuclease | 463.52 |\n",
"| 13 | Darunavir | SARS_CoV2_3to5_exonuclease | 465.23 |\n",
"| 14 | Maraviroc | SARS_CoV2_3to5_exonuclease | 470.40 |\n",
"| 15 | Fosamprenavir | SARS_CoV2_3to5_exonuclease | 479.29 |\n",
"| 16 | Amantadine | SARS_CoV2_3to5_exonuclease | 493.40 |\n",
"| 17 | Efavirenz | SARS_CoV2_3to5_exonuclease | 511.76 |\n",
"| 18 | Elvitegravir | SARS_CoV2_3to5_exonuclease | 546.67 |\n",
"| 19 | Telaprevir | SARS_CoV2_3to5_exonuclease | 553.80 |\n",
"| 20 | Tenofovir | SARS_CoV2_3to5_exonuclease | 566.09 |\n",
"| 21 | Descovy | SARS_CoV2_3to5_exonuclease | 566.09 |\n",
"| 22 | Boceprevir | SARS_CoV2_3to5_exonuclease | 570.51 |\n",
"| 23 | Amprenavir | SARS_CoV2_3to5_exonuclease | 595.71 |\n",
"| 24 | Nelfinavir | SARS_CoV2_3to5_exonuclease | 597.24 |\n",
"| 25 | Doravirine | SARS_CoV2_3to5_exonuclease | 650.27 |\n",
"| 26 | Ritonavir | SARS_CoV2_3to5_exonuclease | 668.31 |\n",
"| 27 | Abacavir | SARS_CoV2_3to5_exonuclease | 720.33 |\n",
"| 28 | Raltegravir | SARS_CoV2_3to5_exonuclease | 771.92 |\n",
"| 29 | Pleconaril | SARS_CoV2_3to5_exonuclease | 830.43 |\n",
"| 30 | Delavirdine | SARS_CoV2_3to5_exonuclease | 864.18 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'agg_mean_max')"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for SARS_CoV2_endoRNAse\n",
"+------+----------------------+---------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+---------------------+---------------+\n",
"| 1 | Daclatasvir | SARS_CoV2_endoRNAse | 425.21 |\n",
"| 2 | Simeprevir | SARS_CoV2_endoRNAse | 464.11 |\n",
"| 3 | Sofosbuvir | SARS_CoV2_endoRNAse | 561.39 |\n",
"| 4 | Vicriviroc | SARS_CoV2_endoRNAse | 794.73 |\n",
"| 5 | Etravirine | SARS_CoV2_endoRNAse | 795.08 |\n",
"| 6 | Atazanavir | SARS_CoV2_endoRNAse | 811.34 |\n",
"| 7 | Rilpivirine | SARS_CoV2_endoRNAse | 869.46 |\n",
"| 8 | Letermovir | SARS_CoV2_endoRNAse | 879.67 |\n",
"| 9 | Maraviroc | SARS_CoV2_endoRNAse | 915.73 |\n",
"| 10 | Darunavir | SARS_CoV2_endoRNAse | 919.07 |\n",
"| 11 | Lopinavir | SARS_CoV2_endoRNAse | 919.69 |\n",
"| 12 | Peramivir | SARS_CoV2_endoRNAse | 939.63 |\n",
"| 13 | Fosamprenavir | SARS_CoV2_endoRNAse | 941.33 |\n",
"| 14 | Grazoprevir | SARS_CoV2_endoRNAse | 1118.11 |\n",
"| 15 | Telaprevir | SARS_CoV2_endoRNAse | 1142.64 |\n",
"| 16 | Amprenavir | SARS_CoV2_endoRNAse | 1176.82 |\n",
"| 17 | Amantadine | SARS_CoV2_endoRNAse | 1190.22 |\n",
"| 18 | Nelfinavir | SARS_CoV2_endoRNAse | 1289.00 |\n",
"| 19 | Elvitegravir | SARS_CoV2_endoRNAse | 1517.18 |\n",
"| 20 | Doravirine | SARS_CoV2_endoRNAse | 1574.81 |\n",
"| 21 | Boceprevir | SARS_CoV2_endoRNAse | 1595.57 |\n",
"| 22 | Raltegravir | SARS_CoV2_endoRNAse | 1661.85 |\n",
"| 23 | Tenofovir_disoproxil | SARS_CoV2_endoRNAse | 1707.86 |\n",
"| 24 | Delavirdine | SARS_CoV2_endoRNAse | 1775.90 |\n",
"| 25 | Abacavir | SARS_CoV2_endoRNAse | 1809.39 |\n",
"| 26 | Saquinavir | SARS_CoV2_endoRNAse | 1812.37 |\n",
"| 27 | Dolutegravir | SARS_CoV2_endoRNAse | 1855.91 |\n",
"| 28 | Ritonavir | SARS_CoV2_endoRNAse | 1902.92 |\n",
"| 29 | Glecaprevir | SARS_CoV2_endoRNAse | 2152.12 |\n",
"| 30 | Pleconaril | SARS_CoV2_endoRNAse | 2189.36 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"target, target_name = dataset.load_SARS_CoV2_endoRNAse()\n",
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'mean')"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for SARS_CoV2_endoRNAse\n",
"+------+----------------------+---------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+---------------------+---------------+\n",
"| 1 | Lopinavir | SARS_CoV2_endoRNAse | 0.18 |\n",
"| 2 | Darunavir | SARS_CoV2_endoRNAse | 0.23 |\n",
"| 3 | Amprenavir | SARS_CoV2_endoRNAse | 0.80 |\n",
"| 4 | Tipranavir | SARS_CoV2_endoRNAse | 2.65 |\n",
"| 5 | Baloxavir | SARS_CoV2_endoRNAse | 3.88 |\n",
"| 6 | Boceprevir | SARS_CoV2_endoRNAse | 3.96 |\n",
"| 7 | Daclatasvir | SARS_CoV2_endoRNAse | 4.57 |\n",
"| 8 | Oseltamivir | SARS_CoV2_endoRNAse | 5.11 |\n",
"| 9 | Vicriviroc | SARS_CoV2_endoRNAse | 5.27 |\n",
"| 10 | Glecaprevir | SARS_CoV2_endoRNAse | 5.33 |\n",
"| 11 | Fosamprenavir | SARS_CoV2_endoRNAse | 5.34 |\n",
"| 12 | Telaprevir | SARS_CoV2_endoRNAse | 5.64 |\n",
"| 13 | Nelfinavir | SARS_CoV2_endoRNAse | 7.91 |\n",
"| 14 | Amantadine | SARS_CoV2_endoRNAse | 7.91 |\n",
"| 15 | Foscarnet | SARS_CoV2_endoRNAse | 10.88 |\n",
"| 16 | Maraviroc | SARS_CoV2_endoRNAse | 12.47 |\n",
"| 17 | Pleconaril | SARS_CoV2_endoRNAse | 13.01 |\n",
"| 18 | Abacavir | SARS_CoV2_endoRNAse | 15.48 |\n",
"| 19 | Sofosbuvir | SARS_CoV2_endoRNAse | 19.61 |\n",
"| 20 | Rimantadine | SARS_CoV2_endoRNAse | 25.77 |\n",
"| 21 | Arbidol | SARS_CoV2_endoRNAse | 26.51 |\n",
"| 22 | Tenofovir | SARS_CoV2_endoRNAse | 29.80 |\n",
"| 23 | Descovy | SARS_CoV2_endoRNAse | 29.80 |\n",
"| 24 | Atazanavir | SARS_CoV2_endoRNAse | 32.23 |\n",
"| 25 | Letermovir | SARS_CoV2_endoRNAse | 32.71 |\n",
"| 26 | Ritonavir | SARS_CoV2_endoRNAse | 35.84 |\n",
"| 27 | Simeprevir | SARS_CoV2_endoRNAse | 36.19 |\n",
"| 28 | Saquinavir | SARS_CoV2_endoRNAse | 37.63 |\n",
"| 29 | Remdesivir | SARS_CoV2_endoRNAse | 38.42 |\n",
"| 30 | Etravirine | SARS_CoV2_endoRNAse | 40.88 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'max_effect')"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for SARS_CoV2_endoRNAse\n",
"+------+----------------------+---------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+---------------------+---------------+\n",
"| 1 | Daclatasvir | SARS_CoV2_endoRNAse | 214.89 |\n",
"| 2 | Simeprevir | SARS_CoV2_endoRNAse | 250.15 |\n",
"| 3 | Sofosbuvir | SARS_CoV2_endoRNAse | 290.50 |\n",
"| 4 | Vicriviroc | SARS_CoV2_endoRNAse | 400.00 |\n",
"| 5 | Etravirine | SARS_CoV2_endoRNAse | 417.98 |\n",
"| 6 | Atazanavir | SARS_CoV2_endoRNAse | 421.78 |\n",
"| 7 | Letermovir | SARS_CoV2_endoRNAse | 456.19 |\n",
"| 8 | Darunavir | SARS_CoV2_endoRNAse | 459.65 |\n",
"| 9 | Lopinavir | SARS_CoV2_endoRNAse | 459.94 |\n",
"| 10 | Maraviroc | SARS_CoV2_endoRNAse | 464.10 |\n",
"| 11 | Rilpivirine | SARS_CoV2_endoRNAse | 468.50 |\n",
"| 12 | Fosamprenavir | SARS_CoV2_endoRNAse | 473.33 |\n",
"| 13 | Peramivir | SARS_CoV2_endoRNAse | 512.31 |\n",
"| 14 | Telaprevir | SARS_CoV2_endoRNAse | 574.14 |\n",
"| 15 | Amprenavir | SARS_CoV2_endoRNAse | 588.81 |\n",
"| 16 | Amantadine | SARS_CoV2_endoRNAse | 599.07 |\n",
"| 17 | Grazoprevir | SARS_CoV2_endoRNAse | 628.48 |\n",
"| 18 | Nelfinavir | SARS_CoV2_endoRNAse | 648.46 |\n",
"| 19 | Boceprevir | SARS_CoV2_endoRNAse | 799.77 |\n",
"| 20 | Elvitegravir | SARS_CoV2_endoRNAse | 817.42 |\n",
"| 21 | Doravirine | SARS_CoV2_endoRNAse | 853.32 |\n",
"| 22 | Raltegravir | SARS_CoV2_endoRNAse | 872.32 |\n",
"| 23 | Abacavir | SARS_CoV2_endoRNAse | 912.43 |\n",
"| 24 | Delavirdine | SARS_CoV2_endoRNAse | 920.16 |\n",
"| 25 | Saquinavir | SARS_CoV2_endoRNAse | 925.00 |\n",
"| 26 | Tenofovir_disoproxil | SARS_CoV2_endoRNAse | 969.32 |\n",
"| 27 | Ritonavir | SARS_CoV2_endoRNAse | 969.38 |\n",
"| 28 | Glecaprevir | SARS_CoV2_endoRNAse | 1078.72 |\n",
"| 29 | Pleconaril | SARS_CoV2_endoRNAse | 1101.19 |\n",
"| 30 | Tenofovir | SARS_CoV2_endoRNAse | 1266.07 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'agg_mean_max')"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for SARS_CoV2_2_O_ribose_methyltransferase\n",
"+------+----------------------+----------------------------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+----------------------------------------+---------------+\n",
"| 1 | Sofosbuvir | SARS_CoV2_2_O_ribose_methyltransferase | 364.51 |\n",
"| 2 | Daclatasvir | SARS_CoV2_2_O_ribose_methyltransferase | 424.59 |\n",
"| 3 | Simeprevir | SARS_CoV2_2_O_ribose_methyltransferase | 512.20 |\n",
"| 4 | Vicriviroc | SARS_CoV2_2_O_ribose_methyltransferase | 739.15 |\n",
"| 5 | Etravirine | SARS_CoV2_2_O_ribose_methyltransferase | 776.94 |\n",
"| 6 | Atazanavir | SARS_CoV2_2_O_ribose_methyltransferase | 835.38 |\n",
"| 7 | Amantadine | SARS_CoV2_2_O_ribose_methyltransferase | 849.80 |\n",
"| 8 | Rilpivirine | SARS_CoV2_2_O_ribose_methyltransferase | 882.95 |\n",
"| 9 | Letermovir | SARS_CoV2_2_O_ribose_methyltransferase | 892.38 |\n",
"| 10 | Ritonavir | SARS_CoV2_2_O_ribose_methyltransferase | 916.95 |\n",
"| 11 | Lopinavir | SARS_CoV2_2_O_ribose_methyltransferase | 953.41 |\n",
"| 12 | Maraviroc | SARS_CoV2_2_O_ribose_methyltransferase | 956.00 |\n",
"| 13 | Darunavir | SARS_CoV2_2_O_ribose_methyltransferase | 956.72 |\n",
"| 14 | Peramivir | SARS_CoV2_2_O_ribose_methyltransferase | 968.37 |\n",
"| 15 | Grazoprevir | SARS_CoV2_2_O_ribose_methyltransferase | 976.05 |\n",
"| 16 | Fosamprenavir | SARS_CoV2_2_O_ribose_methyltransferase | 977.57 |\n",
"| 17 | Efavirenz | SARS_CoV2_2_O_ribose_methyltransferase | 1075.01 |\n",
"| 18 | Telaprevir | SARS_CoV2_2_O_ribose_methyltransferase | 1136.33 |\n",
"| 19 | Elvitegravir | SARS_CoV2_2_O_ribose_methyltransferase | 1188.12 |\n",
"| 20 | Tenofovir | SARS_CoV2_2_O_ribose_methyltransferase | 1200.92 |\n",
"| 21 | Descovy | SARS_CoV2_2_O_ribose_methyltransferase | 1200.92 |\n",
"| 22 | Amprenavir | SARS_CoV2_2_O_ribose_methyltransferase | 1222.05 |\n",
"| 23 | Nelfinavir | SARS_CoV2_2_O_ribose_methyltransferase | 1346.06 |\n",
"| 24 | Tenofovir_disoproxil | SARS_CoV2_2_O_ribose_methyltransferase | 1352.00 |\n",
"| 25 | Tromantadine | SARS_CoV2_2_O_ribose_methyltransferase | 1362.92 |\n",
"| 26 | Doravirine | SARS_CoV2_2_O_ribose_methyltransferase | 1508.91 |\n",
"| 27 | Dolutegravir | SARS_CoV2_2_O_ribose_methyltransferase | 1547.19 |\n",
"| 28 | Abacavir | SARS_CoV2_2_O_ribose_methyltransferase | 1614.96 |\n",
"| 29 | Delavirdine | SARS_CoV2_2_O_ribose_methyltransferase | 1699.89 |\n",
"| 30 | Saquinavir | SARS_CoV2_2_O_ribose_methyltransferase | 1766.76 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"target, target_name = dataset.load_SARS_CoV2_2_O_ribose_methyltransferase()\n",
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'mean')"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for SARS_CoV2_2_O_ribose_methyltransferase\n",
"+------+----------------------+----------------------------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+----------------------------------------+---------------+\n",
"| 1 | Lopinavir | SARS_CoV2_2_O_ribose_methyltransferase | 0.24 |\n",
"| 2 | Darunavir | SARS_CoV2_2_O_ribose_methyltransferase | 0.31 |\n",
"| 3 | Amprenavir | SARS_CoV2_2_O_ribose_methyltransferase | 0.92 |\n",
"| 4 | Tipranavir | SARS_CoV2_2_O_ribose_methyltransferase | 1.46 |\n",
"| 5 | Baloxavir | SARS_CoV2_2_O_ribose_methyltransferase | 1.87 |\n",
"| 6 | Boceprevir | SARS_CoV2_2_O_ribose_methyltransferase | 2.23 |\n",
"| 7 | Glecaprevir | SARS_CoV2_2_O_ribose_methyltransferase | 2.48 |\n",
"| 8 | Oseltamivir | SARS_CoV2_2_O_ribose_methyltransferase | 2.83 |\n",
"| 9 | Telaprevir | SARS_CoV2_2_O_ribose_methyltransferase | 3.00 |\n",
"| 10 | Nelfinavir | SARS_CoV2_2_O_ribose_methyltransferase | 3.35 |\n",
"| 11 | Maraviroc | SARS_CoV2_2_O_ribose_methyltransferase | 4.97 |\n",
"| 12 | Daclatasvir | SARS_CoV2_2_O_ribose_methyltransferase | 5.68 |\n",
"| 13 | Vicriviroc | SARS_CoV2_2_O_ribose_methyltransferase | 6.15 |\n",
"| 14 | Fosamprenavir | SARS_CoV2_2_O_ribose_methyltransferase | 6.23 |\n",
"| 15 | Amantadine | SARS_CoV2_2_O_ribose_methyltransferase | 9.76 |\n",
"| 16 | Etravirine | SARS_CoV2_2_O_ribose_methyltransferase | 10.07 |\n",
"| 17 | Foscarnet | SARS_CoV2_2_O_ribose_methyltransferase | 11.51 |\n",
"| 18 | Atazanavir | SARS_CoV2_2_O_ribose_methyltransferase | 11.70 |\n",
"| 19 | Entecavir | SARS_CoV2_2_O_ribose_methyltransferase | 11.73 |\n",
"| 20 | Pleconaril | SARS_CoV2_2_O_ribose_methyltransferase | 11.91 |\n",
"| 21 | Simeprevir | SARS_CoV2_2_O_ribose_methyltransferase | 13.29 |\n",
"| 22 | Rilpivirine | SARS_CoV2_2_O_ribose_methyltransferase | 13.73 |\n",
"| 23 | Abacavir | SARS_CoV2_2_O_ribose_methyltransferase | 15.62 |\n",
"| 24 | Sofosbuvir | SARS_CoV2_2_O_ribose_methyltransferase | 18.38 |\n",
"| 25 | Saquinavir | SARS_CoV2_2_O_ribose_methyltransferase | 19.33 |\n",
"| 26 | Delavirdine | SARS_CoV2_2_O_ribose_methyltransferase | 20.24 |\n",
"| 27 | Arbidol | SARS_CoV2_2_O_ribose_methyltransferase | 20.65 |\n",
"| 28 | Peramivir | SARS_CoV2_2_O_ribose_methyltransferase | 24.92 |\n",
"| 29 | Raltegravir | SARS_CoV2_2_O_ribose_methyltransferase | 25.47 |\n",
"| 30 | Tenofovir | SARS_CoV2_2_O_ribose_methyltransferase | 25.94 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'max_effect')"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading customized repurposing dataset...\n",
"Checking if pretrained directory is valid...\n",
"Beginning to load the pretrained models...\n",
"Using pretrained model and making predictions...\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 MPNN and target encoding CNN are done...\n",
"-------------\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 CNN and target encoding CNN are done...\n",
"-------------\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 3 with drug encoding Morgan and target encoding CNN are done...\n",
"-------------\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 Morgan and target encoding AAC are done...\n",
"-------------\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 5 with drug encoding Daylight and target encoding AAC are done...\n",
"-------------\n",
"models prediction finished...\n",
"aggregating results...\n",
"---------------\n",
"Drug Repurposing Result for SARS_CoV2_2_O_ribose_methyltransferase\n",
"+------+----------------------+----------------------------------------+---------------+\n",
"| Rank | Drug Name | Target Name | Binding Score |\n",
"+------+----------------------+----------------------------------------+---------------+\n",
"| 1 | Sofosbuvir | SARS_CoV2_2_O_ribose_methyltransferase | 191.45 |\n",
"| 2 | Daclatasvir | SARS_CoV2_2_O_ribose_methyltransferase | 215.14 |\n",
"| 3 | Simeprevir | SARS_CoV2_2_O_ribose_methyltransferase | 262.75 |\n",
"| 4 | Vicriviroc | SARS_CoV2_2_O_ribose_methyltransferase | 372.65 |\n",
"| 5 | Etravirine | SARS_CoV2_2_O_ribose_methyltransferase | 393.50 |\n",
"| 6 | Atazanavir | SARS_CoV2_2_O_ribose_methyltransferase | 423.54 |\n",
"| 7 | Amantadine | SARS_CoV2_2_O_ribose_methyltransferase | 429.78 |\n",
"| 8 | Rilpivirine | SARS_CoV2_2_O_ribose_methyltransferase | 448.34 |\n",
"| 9 | Letermovir | SARS_CoV2_2_O_ribose_methyltransferase | 462.16 |\n",
"| 10 | Lopinavir | SARS_CoV2_2_O_ribose_methyltransferase | 476.83 |\n",
"| 11 | Darunavir | SARS_CoV2_2_O_ribose_methyltransferase | 478.52 |\n",
"| 12 | Ritonavir | SARS_CoV2_2_O_ribose_methyltransferase | 479.50 |\n",
"| 13 | Maraviroc | SARS_CoV2_2_O_ribose_methyltransferase | 480.49 |\n",
"| 14 | Fosamprenavir | SARS_CoV2_2_O_ribose_methyltransferase | 491.90 |\n",
"| 15 | Peramivir | SARS_CoV2_2_O_ribose_methyltransferase | 496.64 |\n",
"| 16 | Grazoprevir | SARS_CoV2_2_O_ribose_methyltransferase | 523.70 |\n",
"| 17 | Telaprevir | SARS_CoV2_2_O_ribose_methyltransferase | 569.67 |\n",
"| 18 | Amprenavir | SARS_CoV2_2_O_ribose_methyltransferase | 611.48 |\n",
"| 19 | Tenofovir | SARS_CoV2_2_O_ribose_methyltransferase | 613.43 |\n",
"| 20 | Descovy | SARS_CoV2_2_O_ribose_methyltransferase | 613.43 |\n",
"| 21 | Elvitegravir | SARS_CoV2_2_O_ribose_methyltransferase | 639.14 |\n",
"| 22 | Efavirenz | SARS_CoV2_2_O_ribose_methyltransferase | 669.05 |\n",
"| 23 | Nelfinavir | SARS_CoV2_2_O_ribose_methyltransferase | 674.70 |\n",
"| 24 | Tenofovir_disoproxil | SARS_CoV2_2_O_ribose_methyltransferase | 719.19 |\n",
"| 25 | Doravirine | SARS_CoV2_2_O_ribose_methyltransferase | 778.37 |\n",
"| 26 | Abacavir | SARS_CoV2_2_O_ribose_methyltransferase | 815.29 |\n",
"| 27 | Delavirdine | SARS_CoV2_2_O_ribose_methyltransferase | 860.06 |\n",
"| 28 | Dolutegravir | SARS_CoV2_2_O_ribose_methyltransferase | 867.61 |\n",
"| 29 | Saquinavir | SARS_CoV2_2_O_ribose_methyltransferase | 893.04 |\n",
"| 30 | Tromantadine | SARS_CoV2_2_O_ribose_methyltransferase | 899.18 |\n",
"checkout ./save_folder/results_aggregation/repurposing.txt for the whole list\n",
"\n"
]
}
],
"source": [
"oneliner.repurpose(target = target, \n",
" target_name = target_name, \n",
" X_repurpose = X_repurpose,\n",
" drug_names = drug_names,\n",
" save_dir = './save_folder',\n",
" pretrained_dir = './save_folder/pretrained_models/DeepPurpose_BindingDB/',\n",
" agg = 'agg_mean_max')"
]
}
],
"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
}