jglaser commited on
Commit
9737168
·
1 Parent(s): 91dc8a9

make smiles canonical

Browse files
Files changed (3) hide show
  1. binding_affinity.py +2 -2
  2. combine_dbs.ipynb +60 -13
  3. data/all.parquet +2 -2
binding_affinity.py CHANGED
@@ -90,7 +90,7 @@ class BindingAffinity(datasets.ArrowBasedBuilder):
90
  features = datasets.Features(
91
  {
92
  "seq": datasets.Value("string"),
93
- "smiles": datasets.Value("string"),
94
  "neg_log10_affinity_M": datasets.Value("float"),
95
  "affinity": datasets.Value("float"),
96
  # These are the features of your dataset like images, labels ...
@@ -161,7 +161,7 @@ class BindingAffinity(datasets.ArrowBasedBuilder):
161
  for k, row in df.iterrows():
162
  yield k, {
163
  "seq": row["seq"],
164
- "smiles": row["smiles"],
165
  "neg_log10_affinity_M": row["neg_log10_affinity_M"],
166
  "affinity_uM": row["affinity_uM"],
167
  }
 
90
  features = datasets.Features(
91
  {
92
  "seq": datasets.Value("string"),
93
+ "smiles_can": datasets.Value("string"),
94
  "neg_log10_affinity_M": datasets.Value("float"),
95
  "affinity": datasets.Value("float"),
96
  # These are the features of your dataset like images, labels ...
 
161
  for k, row in df.iterrows():
162
  yield k, {
163
  "seq": row["seq"],
164
+ "smiles_can": row["smiles_can"],
165
  "neg_log10_affinity_M": row["neg_log10_affinity_M"],
166
  "affinity_uM": row["affinity_uM"],
167
  }
combine_dbs.ipynb CHANGED
@@ -2,7 +2,7 @@
2
  "cells": [
3
  {
4
  "cell_type": "code",
5
- "execution_count": 2,
6
  "id": "95bd761a-fe51-4a8e-bc70-1365260ba5f8",
7
  "metadata": {},
8
  "outputs": [],
@@ -1363,17 +1363,62 @@
1363
  "metadata": {},
1364
  "outputs": [],
1365
  "source": [
1366
- "df.to_parquet('data/all.parquet')"
1367
  ]
1368
  },
1369
  {
1370
  "cell_type": "code",
1371
- "execution_count": 3,
1372
  "id": "4e2d89f7-f6ea-41de-a13b-4a184b4fd580",
1373
  "metadata": {},
1374
  "outputs": [],
1375
  "source": [
1376
- "df = pd.read_parquet('data/all.parquet')"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1377
  ]
1378
  },
1379
  {
@@ -1449,7 +1494,7 @@
1449
  },
1450
  {
1451
  "cell_type": "code",
1452
- "execution_count": 26,
1453
  "id": "c6c64066-4032-4247-a8b9-00388176cc7b",
1454
  "metadata": {},
1455
  "outputs": [],
@@ -1459,17 +1504,19 @@
1459
  },
1460
  {
1461
  "cell_type": "code",
1462
- "execution_count": 9,
1463
  "id": "469cf0dd-7b87-4245-973c-2a445e1fcca9",
1464
  "metadata": {},
1465
  "outputs": [
1466
  {
1467
  "data": {
1468
  "text/plain": [
1469
- "Index(['seq', 'smiles', 'affinity_uM', 'neg_log10_affinity_M', 'affinity'], dtype='object')"
 
 
1470
  ]
1471
  },
1472
- "execution_count": 9,
1473
  "metadata": {},
1474
  "output_type": "execute_result"
1475
  }
@@ -1480,7 +1527,7 @@
1480
  },
1481
  {
1482
  "cell_type": "code",
1483
- "execution_count": 8,
1484
  "id": "d91c0d91-474c-4ab2-9a5e-3b7861f7a832",
1485
  "metadata": {},
1486
  "outputs": [
@@ -1506,23 +1553,23 @@
1506
  },
1507
  {
1508
  "cell_type": "code",
1509
- "execution_count": 16,
1510
  "id": "9ca8df46-15d3-40f9-b304-dd6e5597be5e",
1511
  "metadata": {},
1512
  "outputs": [
1513
  {
1514
  "data": {
1515
  "text/plain": [
1516
- "2938"
1517
  ]
1518
  },
1519
- "execution_count": 16,
1520
  "metadata": {},
1521
  "output_type": "execute_result"
1522
  }
1523
  ],
1524
  "source": [
1525
- "(df['neg_log10_affinity_M']<2).sum()"
1526
  ]
1527
  },
1528
  {
 
2
  "cells": [
3
  {
4
  "cell_type": "code",
5
+ "execution_count": 1,
6
  "id": "95bd761a-fe51-4a8e-bc70-1365260ba5f8",
7
  "metadata": {},
8
  "outputs": [],
 
1363
  "metadata": {},
1364
  "outputs": [],
1365
  "source": [
1366
+ "df.to_parquet('data/all_nocan.parquet')"
1367
  ]
1368
  },
1369
  {
1370
  "cell_type": "code",
1371
+ "execution_count": 11,
1372
  "id": "4e2d89f7-f6ea-41de-a13b-4a184b4fd580",
1373
  "metadata": {},
1374
  "outputs": [],
1375
  "source": [
1376
+ "df = pd.read_parquet('data/all_nocan.parquet')"
1377
+ ]
1378
+ },
1379
+ {
1380
+ "cell_type": "code",
1381
+ "execution_count": 5,
1382
+ "id": "b4b9acd7-7784-492b-9fa3-b7fad9d18a9d",
1383
+ "metadata": {},
1384
+ "outputs": [
1385
+ {
1386
+ "name": "stdout",
1387
+ "output_type": "stream",
1388
+ "text": [
1389
+ "INFO: Pandarallel will run on 256 workers.\n",
1390
+ "INFO: Pandarallel will use Memory file system to transfer data between the main process and workers.\n"
1391
+ ]
1392
+ }
1393
+ ],
1394
+ "source": [
1395
+ "from pandarallel import pandarallel\n",
1396
+ "pandarallel.initialize()\n"
1397
+ ]
1398
+ },
1399
+ {
1400
+ "cell_type": "code",
1401
+ "execution_count": 12,
1402
+ "id": "eb99774f-9bcc-454d-b5e5-a8470223d6ca",
1403
+ "metadata": {},
1404
+ "outputs": [],
1405
+ "source": [
1406
+ "from rdkit import Chem\n",
1407
+ "def make_canonical(smi):\n",
1408
+ " try:\n",
1409
+ " return Chem.MolToSmiles(Chem.MolFromSmiles(smi))\n",
1410
+ " except:\n",
1411
+ " return smi"
1412
+ ]
1413
+ },
1414
+ {
1415
+ "cell_type": "code",
1416
+ "execution_count": 14,
1417
+ "id": "4d44bd8e-f2e1-44b4-aea7-40b4437baf44",
1418
+ "metadata": {},
1419
+ "outputs": [],
1420
+ "source": [
1421
+ "df['smiles_can'] = df['smiles'].parallel_apply(make_canonical)"
1422
  ]
1423
  },
1424
  {
 
1494
  },
1495
  {
1496
  "cell_type": "code",
1497
+ "execution_count": 16,
1498
  "id": "c6c64066-4032-4247-a8b9-00388176cc7b",
1499
  "metadata": {},
1500
  "outputs": [],
 
1504
  },
1505
  {
1506
  "cell_type": "code",
1507
+ "execution_count": 18,
1508
  "id": "469cf0dd-7b87-4245-973c-2a445e1fcca9",
1509
  "metadata": {},
1510
  "outputs": [
1511
  {
1512
  "data": {
1513
  "text/plain": [
1514
+ "Index(['seq', 'smiles', 'affinity_uM', 'neg_log10_affinity_M', 'affinity',\n",
1515
+ " 'smiles_can'],\n",
1516
+ " dtype='object')"
1517
  ]
1518
  },
1519
+ "execution_count": 18,
1520
  "metadata": {},
1521
  "output_type": "execute_result"
1522
  }
 
1527
  },
1528
  {
1529
  "cell_type": "code",
1530
+ "execution_count": 19,
1531
  "id": "d91c0d91-474c-4ab2-9a5e-3b7861f7a832",
1532
  "metadata": {},
1533
  "outputs": [
 
1553
  },
1554
  {
1555
  "cell_type": "code",
1556
+ "execution_count": 6,
1557
  "id": "9ca8df46-15d3-40f9-b304-dd6e5597be5e",
1558
  "metadata": {},
1559
  "outputs": [
1560
  {
1561
  "data": {
1562
  "text/plain": [
1563
+ "0.17005836848632214"
1564
  ]
1565
  },
1566
+ "execution_count": 6,
1567
  "metadata": {},
1568
  "output_type": "execute_result"
1569
  }
1570
  ],
1571
  "source": [
1572
+ "(df['neg_log10_affinity_M']<5).sum()/len(df)"
1573
  ]
1574
  },
1575
  {
data/all.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d22b80e9d2f1c3db65b792395b10f84dfd367dbe908cfd798aea5a10f1c62cb9
3
- size 204641362
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4864292b2aa4e63ffdc28ebdc7baea53a3a396d6e66ccd9927a04885586d160e
3
+ size 228485896