Hugging Face's logo

Datasets:
jglaser
/
binding_affinity

Modalities:
Tabular
Text
Size:
1M - 10M
Tags:
molecules
chemistry
SMILES
Libraries:
Datasets
Dataset card Data Studio Files
xet
Community
2
jglaser commited on
Commit
4e3b836
·
0 Parent(s):

Add README.md, notebooks and preprocessing scripts

Browse files
Files changed (10) hide show
  1. README.md +61 -0
  2. bindingdb.ipynb +791 -0
  3. biolip.ipynb +460 -0
  4. biolip.py +41 -0
  5. combine_dbs.ipynb +1477 -0
  6. moad.ipynb +513 -0
  7. moad.py +32 -0
  8. pdbbind.ipynb +296 -0
  9. pdbbind.py +35 -0
  10. requirements.txt +3 -0
README.md ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ## How to use the data sets
2
+
3
+ ### Use the already preprocessed data
4
+
5
+ The file `data/all.parquet` contains the preprocessed data
6
+
7
+ ### Pre-process yourself
8
+
9
+ To manually perform the preprocessing, fownload the data sets from
10
+
11
+ 1. BindingDB
12
+
13
+ In `bindingdb`, download the database as tab separated values
14
+ [https://bindingdb.org] > Download > BindingDB_All_2021m4.tsv.zip
15
+ and extract the zip archive into `bindingdb/data`
16
+
17
+ Run the steps in `bindingdb.ipynb`
18
+
19
+ 2. PDBBind-cn
20
+
21
+ Register for an account at [https://www.pdbbind.org.cn/], confirm the validation
22
+ email, then login and download
23
+
24
+ - the Index files (1)
25
+ - the general protein-ligand complexes (2)
26
+ - the refined protein-ligand complexes (3)
27
+
28
+ Extract those files in `pdbbind/data`
29
+
30
+ Run the script `pdbbind.py` in a compute job on an MPI-enabled cluster
31
+ (e.g., `mpirun -n 64 pdbbind.py`).
32
+
33
+ Perform the steps in the notebook `pdbbind.ipynb`
34
+
35
+ 3. BindingMOAD
36
+
37
+ Go to [https://bindingmoad.org] and download the files `every.csv`
38
+ (All of Binding MOAD, Binding Data) and the non-redundant biounits
39
+ (`nr_bind.zip`). Place and extract those files into `binding_moad`.
40
+
41
+ Run the script `moad.py` in a compute job on an MPI-enabled cluster
42
+ (e.g., `mpirun -n 64 moad.py).
43
+
44
+ Perform the steps in the notebook `moad.ipynb`
45
+
46
+ 4. BioLIP
47
+
48
+ Download from [https://zhanglab.ccmb.med.umich.edu/BioLiP/] the files
49
+ - receptor_nr1.tar.bz2 (Receptor1, Non-redudant set)
50
+ - ligand_nr.tar.bz2 (Ligands)
51
+ - BioLiP_nr.tar.bz2 (Annotations)
52
+ and extract them in `biolip/data`.
53
+
54
+ Run the script `biolip.py` in a compute job on an MPI-enabled cluster
55
+ (e.g., `mpirun -n 64 biolip.py).
56
+
57
+ Perform sthe steps in the notebook `biolip.ipynb`
58
+
59
+ 5. Final concatenation and filtering
60
+
61
+ Run the steps in the notebook `combine_dbs.ipynb`
bindingdb.ipynb ADDED
@@ -0,0 +1,791 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": 1,
6
+ "id": "ecce356e-321b-441e-8a5d-a20bf72f8691",
7
+ "metadata": {},
8
+ "outputs": [],
9
+ "source": [
10
+ "import dask.dataframe as dd"
11
+ ]
12
+ },
13
+ {
14
+ "cell_type": "code",
15
+ "execution_count": 2,
16
+ "id": "89cbcd82-4ca2-4aba-95b7-e58c0ceed770",
17
+ "metadata": {},
18
+ "outputs": [],
19
+ "source": [
20
+ "cols = ['Ligand SMILES', 'IC50 (nM)','KEGG ID of Ligand','Ki (nM)', 'Kd (nM)','EC50 (nM)']"
21
+ ]
22
+ },
23
+ {
24
+ "cell_type": "code",
25
+ "execution_count": 3,
26
+ "id": "a870d8d7-374b-4474-b9ee-305bbf9f17a9",
27
+ "metadata": {},
28
+ "outputs": [],
29
+ "source": [
30
+ "import tqdm.notebook"
31
+ ]
32
+ },
33
+ {
34
+ "cell_type": "code",
35
+ "execution_count": null,
36
+ "id": "e9f76b32-e8f0-47ee-b592-a91a88f4f93e",
37
+ "metadata": {},
38
+ "outputs": [],
39
+ "source": [
40
+ "for i in tqdm.notebook.tqdm(range(0,13)):\n",
41
+ " mycol = 'BindingDB Target Chain Sequence.{}'.format(i)\n",
42
+ " allseq = ['BindingDB Target Chain Sequence']+['BindingDB Target Chain Sequence.{}'.format(j) for j in range(1,13)]\n",
43
+ " dtypes = {'BindingDB Target Chain Sequence.{}'.format(i): 'object' for i in range(1,13)}\n",
44
+ " dtypes.update({'BindingDB Target Chain Sequence': 'object',\n",
45
+ " 'IC50 (nM)': 'object',\n",
46
+ " 'KEGG ID of Ligand': 'object',\n",
47
+ " 'Ki (nM)': 'object',\n",
48
+ " 'Kd (nM)': 'object',\n",
49
+ " 'EC50 (nM)': 'object',\n",
50
+ " 'koff (s-1)': 'object'})\n",
51
+ " ddf = dd.read_csv('bindingdb/data/BindingDB_All.tsv',sep='\\t',error_bad_lines=False,blocksize=16*1024*1024,\n",
52
+ " usecols=cols+allseq,\n",
53
+ " dtype=dtypes)\n",
54
+ " ddf = ddf.reset_index()\n",
55
+ " ddf = ddf.rename(columns={'BindingDB Target Chain Sequence.{}'.format(j): 'seq_{}'.format(j) for j in range(1,13)})\n",
56
+ " ddf = ddf.rename(columns={'BindingDB Target Chain Sequence': 'seq_0'})\n",
57
+ " ddf = ddf.drop(columns={'seq_{}'.format(j) for j in range(0,13) if i != j})\n",
58
+ " ddf[cols+['seq_{}'.format(i)]].to_parquet('bindingdb/parquet_data/target{}'.format(i),schema='infer')"
59
+ ]
60
+ },
61
+ {
62
+ "cell_type": "code",
63
+ "execution_count": 68,
64
+ "id": "be79bbcf-0622-4d1e-8f08-a723a4167d8b",
65
+ "metadata": {},
66
+ "outputs": [],
67
+ "source": [
68
+ "ddfs = []\n",
69
+ "for i in range(0,13):\n",
70
+ " ddf = dd.read_parquet('bindingdb/parquet_data/target{}'.format(i))\n",
71
+ " ddf = ddf.rename(columns={'seq_{}'.format(i): 'seq'})\n",
72
+ " ddfs.append(ddf)"
73
+ ]
74
+ },
75
+ {
76
+ "cell_type": "code",
77
+ "execution_count": 69,
78
+ "id": "35ca09cb-6264-4526-b504-0d29236a03c1",
79
+ "metadata": {},
80
+ "outputs": [],
81
+ "source": [
82
+ "ddf = dd.concat(ddfs)"
83
+ ]
84
+ },
85
+ {
86
+ "cell_type": "code",
87
+ "execution_count": 70,
88
+ "id": "ba518a9a-0d15-47be-977b-e2dfe2511529",
89
+ "metadata": {},
90
+ "outputs": [
91
+ {
92
+ "data": {
93
+ "text/html": [
94
+ "<div>\n",
95
+ "<style scoped>\n",
96
+ " .dataframe tbody tr th:only-of-type {\n",
97
+ " vertical-align: middle;\n",
98
+ " }\n",
99
+ "\n",
100
+ " .dataframe tbody tr th {\n",
101
+ " vertical-align: top;\n",
102
+ " }\n",
103
+ "\n",
104
+ " .dataframe thead th {\n",
105
+ " text-align: right;\n",
106
+ " }\n",
107
+ "</style>\n",
108
+ "<table border=\"1\" class=\"dataframe\">\n",
109
+ " <thead>\n",
110
+ " <tr style=\"text-align: right;\">\n",
111
+ " <th></th>\n",
112
+ " <th>Ligand SMILES</th>\n",
113
+ " <th>IC50 (nM)</th>\n",
114
+ " <th>KEGG ID of Ligand</th>\n",
115
+ " <th>Ki (nM)</th>\n",
116
+ " <th>Kd (nM)</th>\n",
117
+ " <th>EC50 (nM)</th>\n",
118
+ " <th>seq</th>\n",
119
+ " </tr>\n",
120
+ " </thead>\n",
121
+ " <tbody>\n",
122
+ " <tr>\n",
123
+ " <th>0</th>\n",
124
+ " <td>COc1cc2c(Nc3ccc(Br)cc3F)ncnc2cc1OCC1CCN(C)CC1</td>\n",
125
+ " <td>None</td>\n",
126
+ " <td>None</td>\n",
127
+ " <td>0.24</td>\n",
128
+ " <td>None</td>\n",
129
+ " <td>None</td>\n",
130
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
131
+ " </tr>\n",
132
+ " <tr>\n",
133
+ " <th>1</th>\n",
134
+ " <td>O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(C\\C=C\\c2cn...</td>\n",
135
+ " <td>None</td>\n",
136
+ " <td>None</td>\n",
137
+ " <td>0.25</td>\n",
138
+ " <td>None</td>\n",
139
+ " <td>None</td>\n",
140
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
141
+ " </tr>\n",
142
+ " <tr>\n",
143
+ " <th>2</th>\n",
144
+ " <td>O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(CC2CC2)C(=...</td>\n",
145
+ " <td>None</td>\n",
146
+ " <td>None</td>\n",
147
+ " <td>0.41</td>\n",
148
+ " <td>None</td>\n",
149
+ " <td>None</td>\n",
150
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
151
+ " </tr>\n",
152
+ " <tr>\n",
153
+ " <th>3</th>\n",
154
+ " <td>OCCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@...</td>\n",
155
+ " <td>None</td>\n",
156
+ " <td>None</td>\n",
157
+ " <td>0.8</td>\n",
158
+ " <td>None</td>\n",
159
+ " <td>None</td>\n",
160
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
161
+ " </tr>\n",
162
+ " <tr>\n",
163
+ " <th>4</th>\n",
164
+ " <td>OCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@H...</td>\n",
165
+ " <td>None</td>\n",
166
+ " <td>None</td>\n",
167
+ " <td>0.99</td>\n",
168
+ " <td>None</td>\n",
169
+ " <td>None</td>\n",
170
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
171
+ " </tr>\n",
172
+ " </tbody>\n",
173
+ "</table>\n",
174
+ "</div>"
175
+ ],
176
+ "text/plain": [
177
+ " Ligand SMILES IC50 (nM) \\\n",
178
+ "0 COc1cc2c(Nc3ccc(Br)cc3F)ncnc2cc1OCC1CCN(C)CC1 None \n",
179
+ "1 O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(C\\C=C\\c2cn... None \n",
180
+ "2 O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(CC2CC2)C(=... None \n",
181
+ "3 OCCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@... None \n",
182
+ "4 OCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@H... None \n",
183
+ "\n",
184
+ " KEGG ID of Ligand Ki (nM) Kd (nM) EC50 (nM) \\\n",
185
+ "0 None 0.24 None None \n",
186
+ "1 None 0.25 None None \n",
187
+ "2 None 0.41 None None \n",
188
+ "3 None 0.8 None None \n",
189
+ "4 None 0.99 None None \n",
190
+ "\n",
191
+ " seq \n",
192
+ "0 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... \n",
193
+ "1 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... \n",
194
+ "2 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... \n",
195
+ "3 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... \n",
196
+ "4 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... "
197
+ ]
198
+ },
199
+ "execution_count": 70,
200
+ "metadata": {},
201
+ "output_type": "execute_result"
202
+ }
203
+ ],
204
+ "source": [
205
+ "ddf.head()"
206
+ ]
207
+ },
208
+ {
209
+ "cell_type": "code",
210
+ "execution_count": 71,
211
+ "id": "f504d7aa-dfc1-4346-a136-8814c4b5d979",
212
+ "metadata": {},
213
+ "outputs": [],
214
+ "source": [
215
+ "ddf.repartition(partition_size='25MB').to_parquet('bindingdb/parquet_data/all_targets',schema='infer')"
216
+ ]
217
+ },
218
+ {
219
+ "cell_type": "code",
220
+ "execution_count": 4,
221
+ "id": "d7eafa69-4606-4b34-ae8f-8c6462dcb004",
222
+ "metadata": {},
223
+ "outputs": [],
224
+ "source": [
225
+ "ddf = dd.read_parquet('bindingdb/parquet_data/all_targets')"
226
+ ]
227
+ },
228
+ {
229
+ "cell_type": "code",
230
+ "execution_count": 5,
231
+ "id": "b151868a-0cd6-405e-8401-f79918fb0b07",
232
+ "metadata": {},
233
+ "outputs": [
234
+ {
235
+ "data": {
236
+ "text/html": [
237
+ "<div><strong>Dask DataFrame Structure:</strong></div>\n",
238
+ "<div>\n",
239
+ "<style scoped>\n",
240
+ " .dataframe tbody tr th:only-of-type {\n",
241
+ " vertical-align: middle;\n",
242
+ " }\n",
243
+ "\n",
244
+ " .dataframe tbody tr th {\n",
245
+ " vertical-align: top;\n",
246
+ " }\n",
247
+ "\n",
248
+ " .dataframe thead th {\n",
249
+ " text-align: right;\n",
250
+ " }\n",
251
+ "</style>\n",
252
+ "<table border=\"1\" class=\"dataframe\">\n",
253
+ " <thead>\n",
254
+ " <tr style=\"text-align: right;\">\n",
255
+ " <th></th>\n",
256
+ " <th>Ligand SMILES</th>\n",
257
+ " <th>IC50 (nM)</th>\n",
258
+ " <th>KEGG ID of Ligand</th>\n",
259
+ " <th>Ki (nM)</th>\n",
260
+ " <th>Kd (nM)</th>\n",
261
+ " <th>EC50 (nM)</th>\n",
262
+ " <th>seq</th>\n",
263
+ " </tr>\n",
264
+ " <tr>\n",
265
+ " <th>npartitions=459</th>\n",
266
+ " <th></th>\n",
267
+ " <th></th>\n",
268
+ " <th></th>\n",
269
+ " <th></th>\n",
270
+ " <th></th>\n",
271
+ " <th></th>\n",
272
+ " <th></th>\n",
273
+ " </tr>\n",
274
+ " </thead>\n",
275
+ " <tbody>\n",
276
+ " <tr>\n",
277
+ " <th></th>\n",
278
+ " <td>object</td>\n",
279
+ " <td>object</td>\n",
280
+ " <td>object</td>\n",
281
+ " <td>object</td>\n",
282
+ " <td>object</td>\n",
283
+ " <td>object</td>\n",
284
+ " <td>object</td>\n",
285
+ " </tr>\n",
286
+ " <tr>\n",
287
+ " <th></th>\n",
288
+ " <td>...</td>\n",
289
+ " <td>...</td>\n",
290
+ " <td>...</td>\n",
291
+ " <td>...</td>\n",
292
+ " <td>...</td>\n",
293
+ " <td>...</td>\n",
294
+ " <td>...</td>\n",
295
+ " </tr>\n",
296
+ " <tr>\n",
297
+ " <th>...</th>\n",
298
+ " <td>...</td>\n",
299
+ " <td>...</td>\n",
300
+ " <td>...</td>\n",
301
+ " <td>...</td>\n",
302
+ " <td>...</td>\n",
303
+ " <td>...</td>\n",
304
+ " <td>...</td>\n",
305
+ " </tr>\n",
306
+ " <tr>\n",
307
+ " <th></th>\n",
308
+ " <td>...</td>\n",
309
+ " <td>...</td>\n",
310
+ " <td>...</td>\n",
311
+ " <td>...</td>\n",
312
+ " <td>...</td>\n",
313
+ " <td>...</td>\n",
314
+ " <td>...</td>\n",
315
+ " </tr>\n",
316
+ " <tr>\n",
317
+ " <th></th>\n",
318
+ " <td>...</td>\n",
319
+ " <td>...</td>\n",
320
+ " <td>...</td>\n",
321
+ " <td>...</td>\n",
322
+ " <td>...</td>\n",
323
+ " <td>...</td>\n",
324
+ " <td>...</td>\n",
325
+ " </tr>\n",
326
+ " </tbody>\n",
327
+ "</table>\n",
328
+ "</div>\n",
329
+ "<div>Dask Name: read-parquet, 459 tasks</div>"
330
+ ],
331
+ "text/plain": [
332
+ "Dask DataFrame Structure:\n",
333
+ " Ligand SMILES IC50 (nM) KEGG ID of Ligand Ki (nM) Kd (nM) EC50 (nM) seq\n",
334
+ "npartitions=459 \n",
335
+ " object object object object object object object\n",
336
+ " ... ... ... ... ... ... ...\n",
337
+ "... ... ... ... ... ... ... ...\n",
338
+ " ... ... ... ... ... ... ...\n",
339
+ " ... ... ... ... ... ... ...\n",
340
+ "Dask Name: read-parquet, 459 tasks"
341
+ ]
342
+ },
343
+ "execution_count": 5,
344
+ "metadata": {},
345
+ "output_type": "execute_result"
346
+ }
347
+ ],
348
+ "source": [
349
+ "ddf"
350
+ ]
351
+ },
352
+ {
353
+ "cell_type": "code",
354
+ "execution_count": 6,
355
+ "id": "c00102b8-f4be-4ebd-8d30-7a2c7fc2d05e",
356
+ "metadata": {},
357
+ "outputs": [],
358
+ "source": [
359
+ "ddf_nonnull = ddf[~ddf.seq.isnull()].copy()"
360
+ ]
361
+ },
362
+ {
363
+ "cell_type": "code",
364
+ "execution_count": 7,
365
+ "id": "c5337e06-1e45-4180-90ed-49ac9ecdd24a",
366
+ "metadata": {},
367
+ "outputs": [
368
+ {
369
+ "data": {
370
+ "text/html": [
371
+ "<div>\n",
372
+ "<style scoped>\n",
373
+ " .dataframe tbody tr th:only-of-type {\n",
374
+ " vertical-align: middle;\n",
375
+ " }\n",
376
+ "\n",
377
+ " .dataframe tbody tr th {\n",
378
+ " vertical-align: top;\n",
379
+ " }\n",
380
+ "\n",
381
+ " .dataframe thead th {\n",
382
+ " text-align: right;\n",
383
+ " }\n",
384
+ "</style>\n",
385
+ "<table border=\"1\" class=\"dataframe\">\n",
386
+ " <thead>\n",
387
+ " <tr style=\"text-align: right;\">\n",
388
+ " <th></th>\n",
389
+ " <th>Ligand SMILES</th>\n",
390
+ " <th>IC50 (nM)</th>\n",
391
+ " <th>KEGG ID of Ligand</th>\n",
392
+ " <th>Ki (nM)</th>\n",
393
+ " <th>Kd (nM)</th>\n",
394
+ " <th>EC50 (nM)</th>\n",
395
+ " <th>seq</th>\n",
396
+ " </tr>\n",
397
+ " </thead>\n",
398
+ " <tbody>\n",
399
+ " <tr>\n",
400
+ " <th>4453</th>\n",
401
+ " <td>CC(C)C[C@H](NC(=O)N1CCC(CC1)C(=O)Nc1ccc(cc1)-c...</td>\n",
402
+ " <td>9.4</td>\n",
403
+ " <td>None</td>\n",
404
+ " <td>None</td>\n",
405
+ " <td>None</td>\n",
406
+ " <td>None</td>\n",
407
+ " <td>MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE...</td>\n",
408
+ " </tr>\n",
409
+ " <tr>\n",
410
+ " <th>4454</th>\n",
411
+ " <td>CC(C)C[C@H](NC(=O)[C@H](Cc1ccccc1)NC(=O)c1cncc...</td>\n",
412
+ " <td>11</td>\n",
413
+ " <td>None</td>\n",
414
+ " <td>None</td>\n",
415
+ " <td>None</td>\n",
416
+ " <td>None</td>\n",
417
+ " <td>MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE...</td>\n",
418
+ " </tr>\n",
419
+ " <tr>\n",
420
+ " <th>4455</th>\n",
421
+ " <td>CC(C)C[C@H](NC(=O)N1CCCC(C1)C(=O)Nc1cnccn1)C(=...</td>\n",
422
+ " <td>355</td>\n",
423
+ " <td>None</td>\n",
424
+ " <td>None</td>\n",
425
+ " <td>None</td>\n",
426
+ " <td>None</td>\n",
427
+ " <td>MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE...</td>\n",
428
+ " </tr>\n",
429
+ " <tr>\n",
430
+ " <th>4456</th>\n",
431
+ " <td>COc1ccc(NC(=O)N2CCC(CC2)C(=O)N[C@@H](CC(C)C)C(...</td>\n",
432
+ " <td>17</td>\n",
433
+ " <td>None</td>\n",
434
+ " <td>None</td>\n",
435
+ " <td>None</td>\n",
436
+ " <td>None</td>\n",
437
+ " <td>MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE...</td>\n",
438
+ " </tr>\n",
439
+ " <tr>\n",
440
+ " <th>4457</th>\n",
441
+ " <td>CC(C)C[C@H](NC(=O)C1CCN(CC1)C(=O)Nc1cnccn1)C(=...</td>\n",
442
+ " <td>76</td>\n",
443
+ " <td>None</td>\n",
444
+ " <td>None</td>\n",
445
+ " <td>None</td>\n",
446
+ " <td>None</td>\n",
447
+ " <td>MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE...</td>\n",
448
+ " </tr>\n",
449
+ " </tbody>\n",
450
+ "</table>\n",
451
+ "</div>"
452
+ ],
453
+ "text/plain": [
454
+ " Ligand SMILES IC50 (nM) \\\n",
455
+ "4453 CC(C)C[C@H](NC(=O)N1CCC(CC1)C(=O)Nc1ccc(cc1)-c... 9.4 \n",
456
+ "4454 CC(C)C[C@H](NC(=O)[C@H](Cc1ccccc1)NC(=O)c1cncc... 11 \n",
457
+ "4455 CC(C)C[C@H](NC(=O)N1CCCC(C1)C(=O)Nc1cnccn1)C(=... 355 \n",
458
+ "4456 COc1ccc(NC(=O)N2CCC(CC2)C(=O)N[C@@H](CC(C)C)C(... 17 \n",
459
+ "4457 CC(C)C[C@H](NC(=O)C1CCN(CC1)C(=O)Nc1cnccn1)C(=... 76 \n",
460
+ "\n",
461
+ " KEGG ID of Ligand Ki (nM) Kd (nM) EC50 (nM) \\\n",
462
+ "4453 None None None None \n",
463
+ "4454 None None None None \n",
464
+ "4455 None None None None \n",
465
+ "4456 None None None None \n",
466
+ "4457 None None None None \n",
467
+ "\n",
468
+ " seq \n",
469
+ "4453 MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE... \n",
470
+ "4454 MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE... \n",
471
+ "4455 MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE... \n",
472
+ "4456 MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE... \n",
473
+ "4457 MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE... "
474
+ ]
475
+ },
476
+ "execution_count": 7,
477
+ "metadata": {},
478
+ "output_type": "execute_result"
479
+ }
480
+ ],
481
+ "source": [
482
+ "ddf_nonnull.tail()"
483
+ ]
484
+ },
485
+ {
486
+ "cell_type": "code",
487
+ "execution_count": 8,
488
+ "id": "872edb84-3459-43d9-8e0e-e2a6b5d281eb",
489
+ "metadata": {},
490
+ "outputs": [],
491
+ "source": [
492
+ "from pint import UnitRegistry\n",
493
+ "import numpy as np\n",
494
+ "import re\n",
495
+ "ureg = UnitRegistry()\n",
496
+ "\n",
497
+ "def to_uM(affinities):\n",
498
+ " ic50, Ki, Kd, ec50 = affinities\n",
499
+ "\n",
500
+ " vals = []\n",
501
+ " try:\n",
502
+ " ic50 = ureg(str(ic50)+'nM').m_as(ureg.uM)\n",
503
+ " vals.append(ic50)\n",
504
+ " except:\n",
505
+ " pass\n",
506
+ "\n",
507
+ " try:\n",
508
+ " Ki = ureg(str(Ki)+'nM').m_as(ureg.uM)\n",
509
+ " vals.append(Ki)\n",
510
+ " except:\n",
511
+ " pass\n",
512
+ "\n",
513
+ " try:\n",
514
+ " Kd = ureg(str(Kd)+'nM').m_as(ureg.uM)\n",
515
+ " vals.append(Kd)\n",
516
+ " except:\n",
517
+ " pass\n",
518
+ "\n",
519
+ " try:\n",
520
+ " ec50 = ureg(str(ec50)+'nM').m_as(ureg.uM)\n",
521
+ " vals.append(ec50)\n",
522
+ " except:\n",
523
+ " pass\n",
524
+ "\n",
525
+ " if len(vals) > 0:\n",
526
+ " vals = np.array(vals)\n",
527
+ " return np.mean(vals[~np.isnan(vals)])\n",
528
+ " \n",
529
+ " return None"
530
+ ]
531
+ },
532
+ {
533
+ "cell_type": "code",
534
+ "execution_count": 9,
535
+ "id": "b3cff13c-19b2-4413-a84b-d99062f516a7",
536
+ "metadata": {},
537
+ "outputs": [],
538
+ "source": [
539
+ "df_nonnull = ddf_nonnull.compute()"
540
+ ]
541
+ },
542
+ {
543
+ "cell_type": "code",
544
+ "execution_count": 10,
545
+ "id": "f11834ef-2b8f-4123-816c-5e54ca92a07a",
546
+ "metadata": {},
547
+ "outputs": [
548
+ {
549
+ "name": "stdout",
550
+ "output_type": "stream",
551
+ "text": [
552
+ "Collecting pandarallel\n",
553
+ " Using cached pandarallel-1.5.2.tar.gz (16 kB)\n",
554
+ "Collecting dill\n",
555
+ " Using cached dill-0.3.3-py2.py3-none-any.whl (81 kB)\n",
556
+ "Building wheels for collected packages: pandarallel\n",
557
+ " Building wheel for pandarallel (setup.py) ... \u001b[?25ldone\n",
558
+ "\u001b[?25h Created wheel for pandarallel: filename=pandarallel-1.5.2-py3-none-any.whl size=18384 sha256=d611c0def59d5c3b807ccd787aeba685a821000f283d6082fce6b37d77b4d542\n",
559
+ " Stored in directory: /autofs/nccs-svm1_home1/glaser/.cache/pip/wheels/6e/10/a9/c46b278fe836832830eb22a6a781a8379262d9a82ae87009c1\n",
560
+ "Successfully built pandarallel\n",
561
+ "Installing collected packages: dill, pandarallel\n",
562
+ "Successfully installed dill-0.3.3 pandarallel-1.5.2\n"
563
+ ]
564
+ }
565
+ ],
566
+ "source": [
567
+ "!pip install pandarallel"
568
+ ]
569
+ },
570
+ {
571
+ "cell_type": "code",
572
+ "execution_count": 12,
573
+ "id": "ca9795de-e821-4dc3-a7bf-70ade9e4c7f0",
574
+ "metadata": {},
575
+ "outputs": [
576
+ {
577
+ "name": "stdout",
578
+ "output_type": "stream",
579
+ "text": [
580
+ "INFO: Pandarallel will run on 32 workers.\n",
581
+ "INFO: Pandarallel will use Memory file system to transfer data between the main process and workers.\n"
582
+ ]
583
+ }
584
+ ],
585
+ "source": [
586
+ "from pandarallel import pandarallel\n",
587
+ "pandarallel.initialize()\n"
588
+ ]
589
+ },
590
+ {
591
+ "cell_type": "code",
592
+ "execution_count": 13,
593
+ "id": "4356a3e2-fede-48e7-a486-343661fe0a0a",
594
+ "metadata": {},
595
+ "outputs": [],
596
+ "source": [
597
+ "df_affinity = df_nonnull.copy()\n",
598
+ "df_affinity['affinity_uM'] = df_affinity[['IC50 (nM)', 'Ki (nM)', 'Kd (nM)','EC50 (nM)']].parallel_apply(to_uM,axis=1)"
599
+ ]
600
+ },
601
+ {
602
+ "cell_type": "code",
603
+ "execution_count": 15,
604
+ "id": "e91c3af8-84a5-42a2-9e25-49cb2f320b0b",
605
+ "metadata": {},
606
+ "outputs": [],
607
+ "source": [
608
+ "df_affinity[~df_affinity['affinity_uM'].isnull()].to_parquet('data/bindingdb.parquet')"
609
+ ]
610
+ },
611
+ {
612
+ "cell_type": "code",
613
+ "execution_count": 18,
614
+ "id": "27194288-cf3e-4c30-ad55-3b0998fdf939",
615
+ "metadata": {},
616
+ "outputs": [
617
+ {
618
+ "data": {
619
+ "text/html": [
620
+ "<div>\n",
621
+ "<style scoped>\n",
622
+ " .dataframe tbody tr th:only-of-type {\n",
623
+ " vertical-align: middle;\n",
624
+ " }\n",
625
+ "\n",
626
+ " .dataframe tbody tr th {\n",
627
+ " vertical-align: top;\n",
628
+ " }\n",
629
+ "\n",
630
+ " .dataframe thead th {\n",
631
+ " text-align: right;\n",
632
+ " }\n",
633
+ "</style>\n",
634
+ "<table border=\"1\" class=\"dataframe\">\n",
635
+ " <thead>\n",
636
+ " <tr style=\"text-align: right;\">\n",
637
+ " <th></th>\n",
638
+ " <th>Ligand SMILES</th>\n",
639
+ " <th>IC50 (nM)</th>\n",
640
+ " <th>KEGG ID of Ligand</th>\n",
641
+ " <th>Ki (nM)</th>\n",
642
+ " <th>Kd (nM)</th>\n",
643
+ " <th>EC50 (nM)</th>\n",
644
+ " <th>seq</th>\n",
645
+ " <th>affinity_uM</th>\n",
646
+ " </tr>\n",
647
+ " </thead>\n",
648
+ " <tbody>\n",
649
+ " <tr>\n",
650
+ " <th>0</th>\n",
651
+ " <td>COc1cc2c(Nc3ccc(Br)cc3F)ncnc2cc1OCC1CCN(C)CC1</td>\n",
652
+ " <td>None</td>\n",
653
+ " <td>None</td>\n",
654
+ " <td>0.24</td>\n",
655
+ " <td>None</td>\n",
656
+ " <td>None</td>\n",
657
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
658
+ " <td>0.00024</td>\n",
659
+ " </tr>\n",
660
+ " <tr>\n",
661
+ " <th>1</th>\n",
662
+ " <td>O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(C\\C=C\\c2cn...</td>\n",
663
+ " <td>None</td>\n",
664
+ " <td>None</td>\n",
665
+ " <td>0.25</td>\n",
666
+ " <td>None</td>\n",
667
+ " <td>None</td>\n",
668
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
669
+ " <td>0.00025</td>\n",
670
+ " </tr>\n",
671
+ " <tr>\n",
672
+ " <th>2</th>\n",
673
+ " <td>O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(CC2CC2)C(=...</td>\n",
674
+ " <td>None</td>\n",
675
+ " <td>None</td>\n",
676
+ " <td>0.41</td>\n",
677
+ " <td>None</td>\n",
678
+ " <td>None</td>\n",
679
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
680
+ " <td>0.00041</td>\n",
681
+ " </tr>\n",
682
+ " <tr>\n",
683
+ " <th>3</th>\n",
684
+ " <td>OCCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@...</td>\n",
685
+ " <td>None</td>\n",
686
+ " <td>None</td>\n",
687
+ " <td>0.8</td>\n",
688
+ " <td>None</td>\n",
689
+ " <td>None</td>\n",
690
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
691
+ " <td>0.00080</td>\n",
692
+ " </tr>\n",
693
+ " <tr>\n",
694
+ " <th>4</th>\n",
695
+ " <td>OCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@H...</td>\n",
696
+ " <td>None</td>\n",
697
+ " <td>None</td>\n",
698
+ " <td>0.99</td>\n",
699
+ " <td>None</td>\n",
700
+ " <td>None</td>\n",
701
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
702
+ " <td>0.00099</td>\n",
703
+ " </tr>\n",
704
+ " </tbody>\n",
705
+ "</table>\n",
706
+ "</div>"
707
+ ],
708
+ "text/plain": [
709
+ " Ligand SMILES IC50 (nM) \\\n",
710
+ "0 COc1cc2c(Nc3ccc(Br)cc3F)ncnc2cc1OCC1CCN(C)CC1 None \n",
711
+ "1 O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(C\\C=C\\c2cn... None \n",
712
+ "2 O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(CC2CC2)C(=... None \n",
713
+ "3 OCCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@... None \n",
714
+ "4 OCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@H... None \n",
715
+ "\n",
716
+ " KEGG ID of Ligand Ki (nM) Kd (nM) EC50 (nM) \\\n",
717
+ "0 None 0.24 None None \n",
718
+ "1 None 0.25 None None \n",
719
+ "2 None 0.41 None None \n",
720
+ "3 None 0.8 None None \n",
721
+ "4 None 0.99 None None \n",
722
+ "\n",
723
+ " seq affinity_uM \n",
724
+ "0 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... 0.00024 \n",
725
+ "1 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... 0.00025 \n",
726
+ "2 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... 0.00041 \n",
727
+ "3 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... 0.00080 \n",
728
+ "4 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... 0.00099 "
729
+ ]
730
+ },
731
+ "execution_count": 18,
732
+ "metadata": {},
733
+ "output_type": "execute_result"
734
+ }
735
+ ],
736
+ "source": [
737
+ "df_affinity.head()"
738
+ ]
739
+ },
740
+ {
741
+ "cell_type": "code",
742
+ "execution_count": 17,
743
+ "id": "603fd298-0aa6-4097-b298-c55db013548c",
744
+ "metadata": {},
745
+ "outputs": [
746
+ {
747
+ "data": {
748
+ "text/plain": [
749
+ "2391969"
750
+ ]
751
+ },
752
+ "execution_count": 17,
753
+ "metadata": {},
754
+ "output_type": "execute_result"
755
+ }
756
+ ],
757
+ "source": [
758
+ "len(df_affinity)"
759
+ ]
760
+ },
761
+ {
762
+ "cell_type": "code",
763
+ "execution_count": null,
764
+ "id": "c6ea5a79-facf-4a50-9d7c-2e1864ebad3d",
765
+ "metadata": {},
766
+ "outputs": [],
767
+ "source": []
768
+ }
769
+ ],
770
+ "metadata": {
771
+ "kernelspec": {
772
+ "display_name": "Python 3",
773
+ "language": "python",
774
+ "name": "python3"
775
+ },
776
+ "language_info": {
777
+ "codemirror_mode": {
778
+ "name": "ipython",
779
+ "version": 3
780
+ },
781
+ "file_extension": ".py",
782
+ "mimetype": "text/x-python",
783
+ "name": "python",
784
+ "nbconvert_exporter": "python",
785
+ "pygments_lexer": "ipython3",
786
+ "version": "3.9.4"
787
+ }
788
+ },
789
+ "nbformat": 4,
790
+ "nbformat_minor": 5
791
+ }
biolip.ipynb ADDED
@@ -0,0 +1,460 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": 1,
6
+ "id": "ee43bf48-5491-4dc4-aa09-cb4a0f460f97",
7
+ "metadata": {},
8
+ "outputs": [],
9
+ "source": [
10
+ "from openbabel import pybel\n",
11
+ "from Bio.PDB import * \n"
12
+ ]
13
+ },
14
+ {
15
+ "cell_type": "code",
16
+ "execution_count": 2,
17
+ "id": "26bc18a2-a6eb-49d3-be80-876ddc7dd8e1",
18
+ "metadata": {},
19
+ "outputs": [],
20
+ "source": [
21
+ "import pandas as pd"
22
+ ]
23
+ },
24
+ {
25
+ "cell_type": "code",
26
+ "execution_count": 3,
27
+ "id": "3b59cfb4-c42a-425d-9653-44f07f9e864e",
28
+ "metadata": {},
29
+ "outputs": [],
30
+ "source": [
31
+ "df = pd.read_table('biolip/data/BioLiP_2013-03-6_nr.txt',sep='\\t',header=None,usecols=[0,4,5,6,13,14,15,16,19])\n",
32
+ "df = df.rename(columns={0:'pdb',4:'chain',5:'l_id',6:'l_chain',\n",
33
+ " 13: 'affinity_lit',14: 'affinity_moad',15: 'affinity_pdbbind-cn',16:'affinity_bindingdb',\n",
34
+ " 19: 'seq'})"
35
+ ]
36
+ },
37
+ {
38
+ "cell_type": "code",
39
+ "execution_count": 4,
40
+ "id": "01123edd-2b98-4fcc-a2e9-28213b9bed82",
41
+ "metadata": {},
42
+ "outputs": [],
43
+ "source": [
44
+ "base = 'biolip/data/ligand_nr/'\n",
45
+ "df['ligand_fn'] = base + df['pdb']+'_'+df['chain']+'_'+df['l_id'].astype(str)+'_'+df['l_chain'].astype(str)+'.pdb'"
46
+ ]
47
+ },
48
+ {
49
+ "cell_type": "code",
50
+ "execution_count": 5,
51
+ "id": "bd8671da-66ad-40ad-b221-e33228be65f4",
52
+ "metadata": {},
53
+ "outputs": [],
54
+ "source": [
55
+ "df_complex = pd.read_parquet('data/biolip_complex.parquet')"
56
+ ]
57
+ },
58
+ {
59
+ "cell_type": "code",
60
+ "execution_count": 97,
61
+ "id": "08b04d75-c01e-4b26-ae2d-622efae3bd1f",
62
+ "metadata": {},
63
+ "outputs": [],
64
+ "source": [
65
+ "df_affinity = df_complex[~df_complex['affinity_lit'].isnull() | ~df_complex['affinity_moad'].isnull() \n",
66
+ " | ~df_complex['affinity_pdbbind-cn'].isnull() | ~df_complex['affinity_bindingdb'].isnull()].copy()"
67
+ ]
68
+ },
69
+ {
70
+ "cell_type": "code",
71
+ "execution_count": 98,
72
+ "id": "97af5533-10fe-4419-a998-ed80b7d26690",
73
+ "metadata": {},
74
+ "outputs": [],
75
+ "source": [
76
+ "from pint import UnitRegistry\n",
77
+ "import numpy as np\n",
78
+ "import re\n",
79
+ "ureg = UnitRegistry()\n",
80
+ "\n",
81
+ "def to_uM(affinities):\n",
82
+ " lit, moad, pdbbind, bindingdb = affinities\n",
83
+ "\n",
84
+ " vals = []\n",
85
+ " try:\n",
86
+ " lit = re.split('[=~<>]',str(lit))[1].split(' ')[0]\n",
87
+ " lit = ureg(lit).m_as(ureg.uM)\n",
88
+ " vals.append(lit)\n",
89
+ " except:\n",
90
+ " pass\n",
91
+ "\n",
92
+ " try:\n",
93
+ " moad = re.split('[=~<>]',str(moad))[1].split(' ')[0]\n",
94
+ " moad = ureg(moad).m_as(ureg.uM)\n",
95
+ " vals.append(moad)\n",
96
+ " except:\n",
97
+ " pass\n",
98
+ "\n",
99
+ " try:\n",
100
+ " pdbbind = re.split('[=~<>]',str(pdbbind))[1].split(' ')[0]\n",
101
+ " pdbbind = ureg(bindingdb).m_as(ureg.uM)\n",
102
+ " vals.append(pdbbind)\n",
103
+ " except:\n",
104
+ " pass\n",
105
+ "\n",
106
+ " try:\n",
107
+ " bindingdb = re.split('[=~<>]',str(bindingdb))[1].split(' ')[0]\n",
108
+ " bindingdb = ureg(bindingdb).m_as(ureg.uM)\n",
109
+ " vals.append(bindingdb)\n",
110
+ " except:\n",
111
+ " pass\n",
112
+ "\n",
113
+ " if len(vals) > 0:\n",
114
+ " vals = np.array(vals)\n",
115
+ " return np.mean(vals[~np.isnan(vals)])\n",
116
+ " \n",
117
+ " return None"
118
+ ]
119
+ },
120
+ {
121
+ "cell_type": "code",
122
+ "execution_count": 99,
123
+ "id": "e21154a9-d3a0-4aa3-986f-cfeebc280da6",
124
+ "metadata": {},
125
+ "outputs": [],
126
+ "source": [
127
+ "df_affinity['affinity_uM'] = df_affinity[['affinity_lit','affinity_moad','affinity_pdbbind-cn','affinity_bindingdb']].apply(to_uM,axis=1)"
128
+ ]
129
+ },
130
+ {
131
+ "cell_type": "code",
132
+ "execution_count": 101,
133
+ "id": "0fc94de0-823d-4f4f-9904-1c4d1e722c2e",
134
+ "metadata": {},
135
+ "outputs": [
136
+ {
137
+ "data": {
138
+ "text/html": [
139
+ "<div>\n",
140
+ "<style scoped>\n",
141
+ " .dataframe tbody tr th:only-of-type {\n",
142
+ " vertical-align: middle;\n",
143
+ " }\n",
144
+ "\n",
145
+ " .dataframe tbody tr th {\n",
146
+ " vertical-align: top;\n",
147
+ " }\n",
148
+ "\n",
149
+ " .dataframe thead th {\n",
150
+ " text-align: right;\n",
151
+ " }\n",
152
+ "</style>\n",
153
+ "<table border=\"1\" class=\"dataframe\">\n",
154
+ " <thead>\n",
155
+ " <tr style=\"text-align: right;\">\n",
156
+ " <th></th>\n",
157
+ " <th>pdb</th>\n",
158
+ " <th>chain</th>\n",
159
+ " <th>l_id</th>\n",
160
+ " <th>l_chain</th>\n",
161
+ " <th>affinity_lit</th>\n",
162
+ " <th>affinity_moad</th>\n",
163
+ " <th>affinity_pdbbind-cn</th>\n",
164
+ " <th>affinity_bindingdb</th>\n",
165
+ " <th>seq</th>\n",
166
+ " <th>ligand_fn</th>\n",
167
+ " <th>smiles</th>\n",
168
+ " <th>affinity_uM</th>\n",
169
+ " </tr>\n",
170
+ " </thead>\n",
171
+ " <tbody>\n",
172
+ " <tr>\n",
173
+ " <th>38</th>\n",
174
+ " <td>11gs</td>\n",
175
+ " <td>EAA</td>\n",
176
+ " <td>A</td>\n",
177
+ " <td>1</td>\n",
178
+ " <td>None</td>\n",
179
+ " <td>ki=1.5uM (GTT EAA)</td>\n",
180
+ " <td>Ki=1.5uM (GTT-EAA)</td>\n",
181
+ " <td>None</td>\n",
182
+ " <td>PYTVVYFPVRGRCAALRMLLADQGQSWKEEVVTVETWQEGSLKASC...</td>\n",
183
+ " <td>biolip/data/ligand_nr/11gs_EAA_A_1.pdb</td>\n",
184
+ " <td>CC[C@H](C(=O)c1ccc(c(c1Cl)Cl)OCC(=O)O)C</td>\n",
185
+ " <td>1.500</td>\n",
186
+ " </tr>\n",
187
+ " <tr>\n",
188
+ " <th>43</th>\n",
189
+ " <td>13gs</td>\n",
190
+ " <td>SAS</td>\n",
191
+ " <td>A</td>\n",
192
+ " <td>1</td>\n",
193
+ " <td>None</td>\n",
194
+ " <td>ki=24uM (SAS)</td>\n",
195
+ " <td>Ki=24uM (SAS)</td>\n",
196
+ " <td>None</td>\n",
197
+ " <td>MPPYTVVYFPVRGRCAALRMLLADQGQSWKEEVVTVETWQEGSLKA...</td>\n",
198
+ " <td>biolip/data/ligand_nr/13gs_SAS_A_1.pdb</td>\n",
199
+ " <td>OC(=O)c1cc(/N=N/c2ccc(cc2)S(=O)(=O)Nc2ccccn2)c...</td>\n",
200
+ " <td>24.000</td>\n",
201
+ " </tr>\n",
202
+ " <tr>\n",
203
+ " <th>54</th>\n",
204
+ " <td>17gs</td>\n",
205
+ " <td>GTX</td>\n",
206
+ " <td>A</td>\n",
207
+ " <td>1</td>\n",
208
+ " <td>None</td>\n",
209
+ " <td>None</td>\n",
210
+ " <td>None</td>\n",
211
+ " <td>Kd=10000nM</td>\n",
212
+ " <td>MPPYTVVYFPVRGRCAALRMLLADQGQSWKEEVVTVETWQEGSLKA...</td>\n",
213
+ " <td>biolip/data/ligand_nr/17gs_GTX_A_1.pdb</td>\n",
214
+ " <td>CCCCCCSC[C@@H](C(=O)NCC(=O)O)NC(=O)CC[C@@H](C(...</td>\n",
215
+ " <td>10.000</td>\n",
216
+ " </tr>\n",
217
+ " <tr>\n",
218
+ " <th>55</th>\n",
219
+ " <td>181l</td>\n",
220
+ " <td>BNZ</td>\n",
221
+ " <td>A</td>\n",
222
+ " <td>1</td>\n",
223
+ " <td>None</td>\n",
224
+ " <td>Ka=5700M^-1 (BNZ)</td>\n",
225
+ " <td>None</td>\n",
226
+ " <td>Kd=175000nM</td>\n",
227
+ " <td>MNIFEMLRIDEGLRLKIYKDTEGYYTIGIGHLLTKSPSLNAAKSEL...</td>\n",
228
+ " <td>biolip/data/ligand_nr/181l_BNZ_A_1.pdb</td>\n",
229
+ " <td>c1ccccc1</td>\n",
230
+ " <td>175.000</td>\n",
231
+ " </tr>\n",
232
+ " <tr>\n",
233
+ " <th>56</th>\n",
234
+ " <td>182l</td>\n",
235
+ " <td>BZF</td>\n",
236
+ " <td>A</td>\n",
237
+ " <td>1</td>\n",
238
+ " <td>None</td>\n",
239
+ " <td>Ka=8900M^-1 (BZF)</td>\n",
240
+ " <td>None</td>\n",
241
+ " <td>Kd=112000nM</td>\n",
242
+ " <td>MNIFEMLRIDEGLRLKIYKDTEGYYTIGIGHLLTKSPSLNAAKSEL...</td>\n",
243
+ " <td>biolip/data/ligand_nr/182l_BZF_A_1.pdb</td>\n",
244
+ " <td>c1ccc2c(c1)occ2</td>\n",
245
+ " <td>112.000</td>\n",
246
+ " </tr>\n",
247
+ " <tr>\n",
248
+ " <th>...</th>\n",
249
+ " <td>...</td>\n",
250
+ " <td>...</td>\n",
251
+ " <td>...</td>\n",
252
+ " <td>...</td>\n",
253
+ " <td>...</td>\n",
254
+ " <td>...</td>\n",
255
+ " <td>...</td>\n",
256
+ " <td>...</td>\n",
257
+ " <td>...</td>\n",
258
+ " <td>...</td>\n",
259
+ " <td>...</td>\n",
260
+ " <td>...</td>\n",
261
+ " </tr>\n",
262
+ " <tr>\n",
263
+ " <th>105087</th>\n",
264
+ " <td>8kme</td>\n",
265
+ " <td>III</td>\n",
266
+ " <td>3</td>\n",
267
+ " <td>1</td>\n",
268
+ " <td>None</td>\n",
269
+ " <td>ki=8uM (BNN CUC TRG LEU PRO)</td>\n",
270
+ " <td>None</td>\n",
271
+ " <td>None</td>\n",
272
+ " <td>IVEGSDAEIGMSPWQVMLFRKSPQELLCGASLISDRWVLTAAHCLL...</td>\n",
273
+ " <td>biolip/data/ligand_nr/8kme_III_3_1.pdb</td>\n",
274
+ " <td>O=C[C@@H](NC(=O)[C@H](Cc1ccc(cc1)OS(O)(O)O)NC(...</td>\n",
275
+ " <td>8.000</td>\n",
276
+ " </tr>\n",
277
+ " <tr>\n",
278
+ " <th>105088</th>\n",
279
+ " <td>8kme</td>\n",
280
+ " <td>III</td>\n",
281
+ " <td>4</td>\n",
282
+ " <td>1</td>\n",
283
+ " <td>None</td>\n",
284
+ " <td>ki=8uM (BNN CUC TRG LEU PRO)</td>\n",
285
+ " <td>None</td>\n",
286
+ " <td>None</td>\n",
287
+ " <td>IVEGSDAEIGMSPWQVMLFRKSPQELLCGASLISDRWVLTAAHCLL...</td>\n",
288
+ " <td>biolip/data/ligand_nr/8kme_III_4_1.pdb</td>\n",
289
+ " <td>CC(C[C@@H](C(=O)N1C=CC[C@H]1C(=O)N)NC(=O)[C@@H...</td>\n",
290
+ " <td>8.000</td>\n",
291
+ " </tr>\n",
292
+ " <tr>\n",
293
+ " <th>105106</th>\n",
294
+ " <td>966c</td>\n",
295
+ " <td>RS2</td>\n",
296
+ " <td>A</td>\n",
297
+ " <td>1</td>\n",
298
+ " <td>None</td>\n",
299
+ " <td>ki=23nM (RS2)</td>\n",
300
+ " <td>Ki=23nM (RS2)</td>\n",
301
+ " <td>None</td>\n",
302
+ " <td>RWEQTHLTYRIENYTPDLPRADVDHAIEKAFQLWSNVTPLTFTKVS...</td>\n",
303
+ " <td>biolip/data/ligand_nr/966c_RS2_A_1.pdb</td>\n",
304
+ " <td>ONC(=O)CC1(CCOCC1)S(=O)(=O)c1ccc(cc1)Oc1ccccc1</td>\n",
305
+ " <td>0.023</td>\n",
306
+ " </tr>\n",
307
+ " <tr>\n",
308
+ " <th>105124</th>\n",
309
+ " <td>9icd</td>\n",
310
+ " <td>NAP</td>\n",
311
+ " <td>A</td>\n",
312
+ " <td>1</td>\n",
313
+ " <td>None</td>\n",
314
+ " <td>kd=125uM (NAP)</td>\n",
315
+ " <td>Kd=125uM (NAP)</td>\n",
316
+ " <td>None</td>\n",
317
+ " <td>SKVVVPAQGKKITLQNGKLNVPENPIIPYIEGDGIGVDVTPAMLKV...</td>\n",
318
+ " <td>biolip/data/ligand_nr/9icd_NAP_A_1.pdb</td>\n",
319
+ " <td>O[C@@H]1[C@@H](COP(=O)(O)O)O[C@H]([C@@H]1OP(=O...</td>\n",
320
+ " <td>125.000</td>\n",
321
+ " </tr>\n",
322
+ " <tr>\n",
323
+ " <th>105138</th>\n",
324
+ " <td>9nse</td>\n",
325
+ " <td>ISU</td>\n",
326
+ " <td>B</td>\n",
327
+ " <td>2</td>\n",
328
+ " <td>None</td>\n",
329
+ " <td>Ki=0.039uM (ISU)</td>\n",
330
+ " <td>None</td>\n",
331
+ " <td>None</td>\n",
332
+ " <td>KFPRVKNWELGSITYDTLCAQSQQDGPCTPRRCLGSLVLPRKLQTR...</td>\n",
333
+ " <td>biolip/data/ligand_nr/9nse_ISU_B_2.pdb</td>\n",
334
+ " <td>CC[Se]C(=N)N</td>\n",
335
+ " <td>0.039</td>\n",
336
+ " </tr>\n",
337
+ " </tbody>\n",
338
+ "</table>\n",
339
+ "<p>7588 rows × 12 columns</p>\n",
340
+ "</div>"
341
+ ],
342
+ "text/plain": [
343
+ " pdb chain l_id l_chain affinity_lit affinity_moad \\\n",
344
+ "38 11gs EAA A 1 None ki=1.5uM (GTT EAA) \n",
345
+ "43 13gs SAS A 1 None ki=24uM (SAS) \n",
346
+ "54 17gs GTX A 1 None None \n",
347
+ "55 181l BNZ A 1 None Ka=5700M^-1 (BNZ) \n",
348
+ "56 182l BZF A 1 None Ka=8900M^-1 (BZF) \n",
349
+ "... ... ... ... ... ... ... \n",
350
+ "105087 8kme III 3 1 None ki=8uM (BNN CUC TRG LEU PRO) \n",
351
+ "105088 8kme III 4 1 None ki=8uM (BNN CUC TRG LEU PRO) \n",
352
+ "105106 966c RS2 A 1 None ki=23nM (RS2) \n",
353
+ "105124 9icd NAP A 1 None kd=125uM (NAP) \n",
354
+ "105138 9nse ISU B 2 None Ki=0.039uM (ISU) \n",
355
+ "\n",
356
+ " affinity_pdbbind-cn affinity_bindingdb \\\n",
357
+ "38 Ki=1.5uM (GTT-EAA) None \n",
358
+ "43 Ki=24uM (SAS) None \n",
359
+ "54 None Kd=10000nM \n",
360
+ "55 None Kd=175000nM \n",
361
+ "56 None Kd=112000nM \n",
362
+ "... ... ... \n",
363
+ "105087 None None \n",
364
+ "105088 None None \n",
365
+ "105106 Ki=23nM (RS2) None \n",
366
+ "105124 Kd=125uM (NAP) None \n",
367
+ "105138 None None \n",
368
+ "\n",
369
+ " seq \\\n",
370
+ "38 PYTVVYFPVRGRCAALRMLLADQGQSWKEEVVTVETWQEGSLKASC... \n",
371
+ "43 MPPYTVVYFPVRGRCAALRMLLADQGQSWKEEVVTVETWQEGSLKA... \n",
372
+ "54 MPPYTVVYFPVRGRCAALRMLLADQGQSWKEEVVTVETWQEGSLKA... \n",
373
+ "55 MNIFEMLRIDEGLRLKIYKDTEGYYTIGIGHLLTKSPSLNAAKSEL... \n",
374
+ "56 MNIFEMLRIDEGLRLKIYKDTEGYYTIGIGHLLTKSPSLNAAKSEL... \n",
375
+ "... ... \n",
376
+ "105087 IVEGSDAEIGMSPWQVMLFRKSPQELLCGASLISDRWVLTAAHCLL... \n",
377
+ "105088 IVEGSDAEIGMSPWQVMLFRKSPQELLCGASLISDRWVLTAAHCLL... \n",
378
+ "105106 RWEQTHLTYRIENYTPDLPRADVDHAIEKAFQLWSNVTPLTFTKVS... \n",
379
+ "105124 SKVVVPAQGKKITLQNGKLNVPENPIIPYIEGDGIGVDVTPAMLKV... \n",
380
+ "105138 KFPRVKNWELGSITYDTLCAQSQQDGPCTPRRCLGSLVLPRKLQTR... \n",
381
+ "\n",
382
+ " ligand_fn \\\n",
383
+ "38 biolip/data/ligand_nr/11gs_EAA_A_1.pdb \n",
384
+ "43 biolip/data/ligand_nr/13gs_SAS_A_1.pdb \n",
385
+ "54 biolip/data/ligand_nr/17gs_GTX_A_1.pdb \n",
386
+ "55 biolip/data/ligand_nr/181l_BNZ_A_1.pdb \n",
387
+ "56 biolip/data/ligand_nr/182l_BZF_A_1.pdb \n",
388
+ "... ... \n",
389
+ "105087 biolip/data/ligand_nr/8kme_III_3_1.pdb \n",
390
+ "105088 biolip/data/ligand_nr/8kme_III_4_1.pdb \n",
391
+ "105106 biolip/data/ligand_nr/966c_RS2_A_1.pdb \n",
392
+ "105124 biolip/data/ligand_nr/9icd_NAP_A_1.pdb \n",
393
+ "105138 biolip/data/ligand_nr/9nse_ISU_B_2.pdb \n",
394
+ "\n",
395
+ " smiles affinity_uM \n",
396
+ "38 CC[C@H](C(=O)c1ccc(c(c1Cl)Cl)OCC(=O)O)C 1.500 \n",
397
+ "43 OC(=O)c1cc(/N=N/c2ccc(cc2)S(=O)(=O)Nc2ccccn2)c... 24.000 \n",
398
+ "54 CCCCCCSC[C@@H](C(=O)NCC(=O)O)NC(=O)CC[C@@H](C(... 10.000 \n",
399
+ "55 c1ccccc1 175.000 \n",
400
+ "56 c1ccc2c(c1)occ2 112.000 \n",
401
+ "... ... ... \n",
402
+ "105087 O=C[C@@H](NC(=O)[C@H](Cc1ccc(cc1)OS(O)(O)O)NC(... 8.000 \n",
403
+ "105088 CC(C[C@@H](C(=O)N1C=CC[C@H]1C(=O)N)NC(=O)[C@@H... 8.000 \n",
404
+ "105106 ONC(=O)CC1(CCOCC1)S(=O)(=O)c1ccc(cc1)Oc1ccccc1 0.023 \n",
405
+ "105124 O[C@@H]1[C@@H](COP(=O)(O)O)O[C@H]([C@@H]1OP(=O... 125.000 \n",
406
+ "105138 CC[Se]C(=N)N 0.039 \n",
407
+ "\n",
408
+ "[7588 rows x 12 columns]"
409
+ ]
410
+ },
411
+ "execution_count": 101,
412
+ "metadata": {},
413
+ "output_type": "execute_result"
414
+ }
415
+ ],
416
+ "source": [
417
+ "df_affinity[~df_affinity['affinity_uM'].isnull()]"
418
+ ]
419
+ },
420
+ {
421
+ "cell_type": "code",
422
+ "execution_count": 102,
423
+ "id": "2b483565-3c99-4c42-b2a9-f7b97cd8e80e",
424
+ "metadata": {},
425
+ "outputs": [],
426
+ "source": [
427
+ "df_affinity.to_parquet('data/biolip.parquet')"
428
+ ]
429
+ },
430
+ {
431
+ "cell_type": "code",
432
+ "execution_count": null,
433
+ "id": "68dd5e45-b31d-492d-a47e-39072b67fa72",
434
+ "metadata": {},
435
+ "outputs": [],
436
+ "source": []
437
+ }
438
+ ],
439
+ "metadata": {
440
+ "kernelspec": {
441
+ "display_name": "Python 3",
442
+ "language": "python",
443
+ "name": "python3"
444
+ },
445
+ "language_info": {
446
+ "codemirror_mode": {
447
+ "name": "ipython",
448
+ "version": 3
449
+ },
450
+ "file_extension": ".py",
451
+ "mimetype": "text/x-python",
452
+ "name": "python",
453
+ "nbconvert_exporter": "python",
454
+ "pygments_lexer": "ipython3",
455
+ "version": "3.9.4"
456
+ }
457
+ },
458
+ "nbformat": 4,
459
+ "nbformat_minor": 5
460
+ }
biolip.py ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from mpi4py import MPI
2
+ from mpi4py.futures import MPICommExecutor
3
+
4
+ from openbabel import pybel
5
+ from Bio.PDB import *
6
+ parser = PDBParser()
7
+
8
+ import os
9
+ molecular_weight_cutoff = 2500
10
+ def parse_ligand(fn):
11
+ print(fn)
12
+ try:
13
+ struct = parser.get_structure('lig',fn)
14
+ if len(list(struct.get_atoms())) > molecular_weight_cutoff:
15
+ raise ValueError
16
+ mol = next(pybel.readfile('pdb',fn))
17
+ if mol.molwt > molecular_weight_cutoff:
18
+ raise ValueError
19
+ smi = mol.write('can').split('\t')[0]
20
+ return smi
21
+ except:
22
+ return None
23
+
24
+
25
+ if __name__ == '__main__':
26
+ import glob
27
+
28
+ comm = MPI.COMM_WORLD
29
+ with MPICommExecutor(comm, root=0) as executor:
30
+ if executor is not None:
31
+ import pandas as pd
32
+
33
+ df = pd.read_table('biolip/data/BioLiP_2013-03-6_nr.txt',sep='\t',header=None,usecols=[0,4,5,6,13,14,15,16,19])
34
+ df = df.rename(columns={0:'pdb',4:'chain',5:'l_id',6:'l_chain',
35
+ 13: 'affinity_lit',14: 'affinity_moad',15: 'affinity_pdbbind-cn',16:'affinity_bindingdb',
36
+ 19: 'seq'})
37
+ base = 'biolip/data/ligand_nr/'
38
+ df['ligand_fn'] = base + df['pdb']+'_'+df['chain']+'_'+df['l_id'].astype(str)+'_'+df['l_chain'].astype(str)+'.pdb'
39
+ smiles = list(executor.map(parse_ligand, df['ligand_fn']))
40
+ df['smiles'] = smiles
41
+ df.to_parquet('data/biolip_complex.parquet')
combine_dbs.ipynb ADDED
@@ -0,0 +1,1477 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": 2,
6
+ "id": "95bd761a-fe51-4a8e-bc70-1365260ba5f8",
7
+ "metadata": {},
8
+ "outputs": [],
9
+ "source": [
10
+ "import pandas as pd"
11
+ ]
12
+ },
13
+ {
14
+ "cell_type": "code",
15
+ "execution_count": 84,
16
+ "id": "b0859483-5e19-4280-9f53-0d00a6f22d34",
17
+ "metadata": {},
18
+ "outputs": [],
19
+ "source": [
20
+ "df_pdbbind = pd.read_parquet('data/pdbbind.parquet')\n",
21
+ "df_pdbbind = df_pdbbind[['seq','smiles','affinity_uM']]"
22
+ ]
23
+ },
24
+ {
25
+ "cell_type": "code",
26
+ "execution_count": 85,
27
+ "id": "f30732b7-7444-47ad-84e7-566e7a6f2f8e",
28
+ "metadata": {},
29
+ "outputs": [
30
+ {
31
+ "data": {
32
+ "text/html": [
33
+ "<div>\n",
34
+ "<style scoped>\n",
35
+ " .dataframe tbody tr th:only-of-type {\n",
36
+ " vertical-align: middle;\n",
37
+ " }\n",
38
+ "\n",
39
+ " .dataframe tbody tr th {\n",
40
+ " vertical-align: top;\n",
41
+ " }\n",
42
+ "\n",
43
+ " .dataframe thead th {\n",
44
+ " text-align: right;\n",
45
+ " }\n",
46
+ "</style>\n",
47
+ "<table border=\"1\" class=\"dataframe\">\n",
48
+ " <thead>\n",
49
+ " <tr style=\"text-align: right;\">\n",
50
+ " <th></th>\n",
51
+ " <th>seq</th>\n",
52
+ " <th>smiles</th>\n",
53
+ " <th>affinity_uM</th>\n",
54
+ " </tr>\n",
55
+ " </thead>\n",
56
+ " <tbody>\n",
57
+ " <tr>\n",
58
+ " <th>0</th>\n",
59
+ " <td>MTVPDRSEIAGKWYVVALASNTEFFLREKDKMKMAMARISFLGEDE...</td>\n",
60
+ " <td>CCCCCCCCCCCCCCCCCCC[C-](=O)=O</td>\n",
61
+ " <td>0.026</td>\n",
62
+ " </tr>\n",
63
+ " <tr>\n",
64
+ " <th>1</th>\n",
65
+ " <td>APQTITELCSEYRNTQIYTINDKILSYTESMAGKREMVIITFKSGE...</td>\n",
66
+ " <td>OC[C@H]1O[C@H](Oc2cccc(c2)N(=O)=O)[C@@H]([C@H]...</td>\n",
67
+ " <td>500.000</td>\n",
68
+ " </tr>\n",
69
+ " <tr>\n",
70
+ " <th>2</th>\n",
71
+ " <td>VETFAFQAEIAQLMSLIINTFYSNKEIFLRELISNSSDALDKIRYE...</td>\n",
72
+ " <td>COc1ccc(cc1)c1c(onc1c1cc(C(C)C)c(cc1O)O)NC(=O)...</td>\n",
73
+ " <td>0.023</td>\n",
74
+ " </tr>\n",
75
+ " <tr>\n",
76
+ " <th>3</th>\n",
77
+ " <td>AAPFDKSKNVAQSIDQLIGQTPALYLNKLNNTKAKVVLKMECENPM...</td>\n",
78
+ " <td>OC[C@@H](C(=O)N[C@@H]([C@H](CC)C)[C-](=O)=O)NC...</td>\n",
79
+ " <td>6.430</td>\n",
80
+ " </tr>\n",
81
+ " <tr>\n",
82
+ " <th>4</th>\n",
83
+ " <td>GSFVEMVDNLRGKSGQGYYVEMTVGSPPQTLNILVDTGSSNFAVGA...</td>\n",
84
+ " <td>O=[C-](=O)[C@@H](NC1=NC(C)(C)Cc2c1cccc2)Cc1ccccc1</td>\n",
85
+ " <td>27.200</td>\n",
86
+ " </tr>\n",
87
+ " </tbody>\n",
88
+ "</table>\n",
89
+ "</div>"
90
+ ],
91
+ "text/plain": [
92
+ " seq \\\n",
93
+ "0 MTVPDRSEIAGKWYVVALASNTEFFLREKDKMKMAMARISFLGEDE... \n",
94
+ "1 APQTITELCSEYRNTQIYTINDKILSYTESMAGKREMVIITFKSGE... \n",
95
+ "2 VETFAFQAEIAQLMSLIINTFYSNKEIFLRELISNSSDALDKIRYE... \n",
96
+ "3 AAPFDKSKNVAQSIDQLIGQTPALYLNKLNNTKAKVVLKMECENPM... \n",
97
+ "4 GSFVEMVDNLRGKSGQGYYVEMTVGSPPQTLNILVDTGSSNFAVGA... \n",
98
+ "\n",
99
+ " smiles affinity_uM \n",
100
+ "0 CCCCCCCCCCCCCCCCCCC[C-](=O)=O 0.026 \n",
101
+ "1 OC[C@H]1O[C@H](Oc2cccc(c2)N(=O)=O)[C@@H]([C@H]... 500.000 \n",
102
+ "2 COc1ccc(cc1)c1c(onc1c1cc(C(C)C)c(cc1O)O)NC(=O)... 0.023 \n",
103
+ "3 OC[C@@H](C(=O)N[C@@H]([C@H](CC)C)[C-](=O)=O)NC... 6.430 \n",
104
+ "4 O=[C-](=O)[C@@H](NC1=NC(C)(C)Cc2c1cccc2)Cc1ccccc1 27.200 "
105
+ ]
106
+ },
107
+ "execution_count": 85,
108
+ "metadata": {},
109
+ "output_type": "execute_result"
110
+ }
111
+ ],
112
+ "source": [
113
+ "df_pdbbind.head()"
114
+ ]
115
+ },
116
+ {
117
+ "cell_type": "code",
118
+ "execution_count": 119,
119
+ "id": "2787b9fd-3d6f-4ae3-a3ad-d3539b72782b",
120
+ "metadata": {},
121
+ "outputs": [],
122
+ "source": [
123
+ "from rdkit import Chem\n",
124
+ "from rdkit.Chem import MACCSkeys\n",
125
+ "import numpy as np\n",
126
+ "\n",
127
+ "def get_maccs(smi):\n",
128
+ " try:\n",
129
+ " mol = Chem.MolFromSmiles(smi)\n",
130
+ " arr = np.packbits([0 if c=='0' else 1 for c in MACCSkeys.GenMACCSKeys(mol).ToBitString()])\n",
131
+ " return np.pad(arr,(0,3)).view(np.uint32)\n",
132
+ " except Exception:\n",
133
+ " pass"
134
+ ]
135
+ },
136
+ {
137
+ "cell_type": "code",
138
+ "execution_count": 120,
139
+ "id": "84f522d5-aee8-4d0f-9186-2d90bfc62342",
140
+ "metadata": {},
141
+ "outputs": [
142
+ {
143
+ "data": {
144
+ "text/html": [
145
+ "<div>\n",
146
+ "<style scoped>\n",
147
+ " .dataframe tbody tr th:only-of-type {\n",
148
+ " vertical-align: middle;\n",
149
+ " }\n",
150
+ "\n",
151
+ " .dataframe tbody tr th {\n",
152
+ " vertical-align: top;\n",
153
+ " }\n",
154
+ "\n",
155
+ " .dataframe thead th {\n",
156
+ " text-align: right;\n",
157
+ " }\n",
158
+ "</style>\n",
159
+ "<table border=\"1\" class=\"dataframe\">\n",
160
+ " <thead>\n",
161
+ " <tr style=\"text-align: right;\">\n",
162
+ " <th></th>\n",
163
+ " <th>seq</th>\n",
164
+ " <th>smiles</th>\n",
165
+ " <th>affinity_uM</th>\n",
166
+ " </tr>\n",
167
+ " </thead>\n",
168
+ " <tbody>\n",
169
+ " <tr>\n",
170
+ " <th>0</th>\n",
171
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
172
+ " <td>COc1cc2c(Nc3ccc(Br)cc3F)ncnc2cc1OCC1CCN(C)CC1</td>\n",
173
+ " <td>0.00024</td>\n",
174
+ " </tr>\n",
175
+ " <tr>\n",
176
+ " <th>1</th>\n",
177
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
178
+ " <td>O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(C\\C=C\\c2cn...</td>\n",
179
+ " <td>0.00025</td>\n",
180
+ " </tr>\n",
181
+ " <tr>\n",
182
+ " <th>2</th>\n",
183
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
184
+ " <td>O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(CC2CC2)C(=...</td>\n",
185
+ " <td>0.00041</td>\n",
186
+ " </tr>\n",
187
+ " <tr>\n",
188
+ " <th>3</th>\n",
189
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
190
+ " <td>OCCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@...</td>\n",
191
+ " <td>0.00080</td>\n",
192
+ " </tr>\n",
193
+ " <tr>\n",
194
+ " <th>4</th>\n",
195
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
196
+ " <td>OCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@H...</td>\n",
197
+ " <td>0.00099</td>\n",
198
+ " </tr>\n",
199
+ " <tr>\n",
200
+ " <th>...</th>\n",
201
+ " <td>...</td>\n",
202
+ " <td>...</td>\n",
203
+ " <td>...</td>\n",
204
+ " </tr>\n",
205
+ " <tr>\n",
206
+ " <th>4453</th>\n",
207
+ " <td>MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE...</td>\n",
208
+ " <td>CC(C)C[C@H](NC(=O)N1CCC(CC1)C(=O)Nc1ccc(cc1)-c...</td>\n",
209
+ " <td>0.00940</td>\n",
210
+ " </tr>\n",
211
+ " <tr>\n",
212
+ " <th>4454</th>\n",
213
+ " <td>MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE...</td>\n",
214
+ " <td>CC(C)C[C@H](NC(=O)[C@H](Cc1ccccc1)NC(=O)c1cncc...</td>\n",
215
+ " <td>0.01100</td>\n",
216
+ " </tr>\n",
217
+ " <tr>\n",
218
+ " <th>4455</th>\n",
219
+ " <td>MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE...</td>\n",
220
+ " <td>CC(C)C[C@H](NC(=O)N1CCCC(C1)C(=O)Nc1cnccn1)C(=...</td>\n",
221
+ " <td>0.35500</td>\n",
222
+ " </tr>\n",
223
+ " <tr>\n",
224
+ " <th>4456</th>\n",
225
+ " <td>MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE...</td>\n",
226
+ " <td>COc1ccc(NC(=O)N2CCC(CC2)C(=O)N[C@@H](CC(C)C)C(...</td>\n",
227
+ " <td>0.01700</td>\n",
228
+ " </tr>\n",
229
+ " <tr>\n",
230
+ " <th>4457</th>\n",
231
+ " <td>MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE...</td>\n",
232
+ " <td>CC(C)C[C@H](NC(=O)C1CCN(CC1)C(=O)Nc1cnccn1)C(=...</td>\n",
233
+ " <td>0.07600</td>\n",
234
+ " </tr>\n",
235
+ " </tbody>\n",
236
+ "</table>\n",
237
+ "<p>2389700 rows × 3 columns</p>\n",
238
+ "</div>"
239
+ ],
240
+ "text/plain": [
241
+ " seq \\\n",
242
+ "0 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... \n",
243
+ "1 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... \n",
244
+ "2 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... \n",
245
+ "3 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... \n",
246
+ "4 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... \n",
247
+ "... ... \n",
248
+ "4453 MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE... \n",
249
+ "4454 MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE... \n",
250
+ "4455 MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE... \n",
251
+ "4456 MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE... \n",
252
+ "4457 MSYDRAITVFSPDGHLFQVEYAQEAVKKGSTAVGVRGRDIVVLGVE... \n",
253
+ "\n",
254
+ " smiles affinity_uM \n",
255
+ "0 COc1cc2c(Nc3ccc(Br)cc3F)ncnc2cc1OCC1CCN(C)CC1 0.00024 \n",
256
+ "1 O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(C\\C=C\\c2cn... 0.00025 \n",
257
+ "2 O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(CC2CC2)C(=... 0.00041 \n",
258
+ "3 OCCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@... 0.00080 \n",
259
+ "4 OCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@H... 0.00099 \n",
260
+ "... ... ... \n",
261
+ "4453 CC(C)C[C@H](NC(=O)N1CCC(CC1)C(=O)Nc1ccc(cc1)-c... 0.00940 \n",
262
+ "4454 CC(C)C[C@H](NC(=O)[C@H](Cc1ccccc1)NC(=O)c1cncc... 0.01100 \n",
263
+ "4455 CC(C)C[C@H](NC(=O)N1CCCC(C1)C(=O)Nc1cnccn1)C(=... 0.35500 \n",
264
+ "4456 COc1ccc(NC(=O)N2CCC(CC2)C(=O)N[C@@H](CC(C)C)C(... 0.01700 \n",
265
+ "4457 CC(C)C[C@H](NC(=O)C1CCN(CC1)C(=O)Nc1cnccn1)C(=... 0.07600 \n",
266
+ "\n",
267
+ "[2389700 rows x 3 columns]"
268
+ ]
269
+ },
270
+ "execution_count": 120,
271
+ "metadata": {},
272
+ "output_type": "execute_result"
273
+ }
274
+ ],
275
+ "source": [
276
+ "df_bindingdb"
277
+ ]
278
+ },
279
+ {
280
+ "cell_type": "code",
281
+ "execution_count": 88,
282
+ "id": "d1abe1c8-ac66-4289-8964-367a5b18528d",
283
+ "metadata": {},
284
+ "outputs": [],
285
+ "source": [
286
+ "df_bindingdb = pd.read_parquet('data/bindingdb.parquet')\n",
287
+ "df_bindingdb = df_bindingdb[['seq','Ligand SMILES','affinity_uM']].rename(columns={'Ligand SMILES': 'smiles'})"
288
+ ]
289
+ },
290
+ {
291
+ "cell_type": "code",
292
+ "execution_count": 89,
293
+ "id": "988bab9c-5147-44e2-92ef-902eaf3c5a90",
294
+ "metadata": {},
295
+ "outputs": [
296
+ {
297
+ "data": {
298
+ "text/html": [
299
+ "<div>\n",
300
+ "<style scoped>\n",
301
+ " .dataframe tbody tr th:only-of-type {\n",
302
+ " vertical-align: middle;\n",
303
+ " }\n",
304
+ "\n",
305
+ " .dataframe tbody tr th {\n",
306
+ " vertical-align: top;\n",
307
+ " }\n",
308
+ "\n",
309
+ " .dataframe thead th {\n",
310
+ " text-align: right;\n",
311
+ " }\n",
312
+ "</style>\n",
313
+ "<table border=\"1\" class=\"dataframe\">\n",
314
+ " <thead>\n",
315
+ " <tr style=\"text-align: right;\">\n",
316
+ " <th></th>\n",
317
+ " <th>seq</th>\n",
318
+ " <th>smiles</th>\n",
319
+ " <th>affinity_uM</th>\n",
320
+ " </tr>\n",
321
+ " </thead>\n",
322
+ " <tbody>\n",
323
+ " <tr>\n",
324
+ " <th>0</th>\n",
325
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
326
+ " <td>COc1cc2c(Nc3ccc(Br)cc3F)ncnc2cc1OCC1CCN(C)CC1</td>\n",
327
+ " <td>0.00024</td>\n",
328
+ " </tr>\n",
329
+ " <tr>\n",
330
+ " <th>1</th>\n",
331
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
332
+ " <td>O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(C\\C=C\\c2cn...</td>\n",
333
+ " <td>0.00025</td>\n",
334
+ " </tr>\n",
335
+ " <tr>\n",
336
+ " <th>2</th>\n",
337
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
338
+ " <td>O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(CC2CC2)C(=...</td>\n",
339
+ " <td>0.00041</td>\n",
340
+ " </tr>\n",
341
+ " <tr>\n",
342
+ " <th>3</th>\n",
343
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
344
+ " <td>OCCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@...</td>\n",
345
+ " <td>0.00080</td>\n",
346
+ " </tr>\n",
347
+ " <tr>\n",
348
+ " <th>4</th>\n",
349
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM...</td>\n",
350
+ " <td>OCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@H...</td>\n",
351
+ " <td>0.00099</td>\n",
352
+ " </tr>\n",
353
+ " </tbody>\n",
354
+ "</table>\n",
355
+ "</div>"
356
+ ],
357
+ "text/plain": [
358
+ " seq \\\n",
359
+ "0 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... \n",
360
+ "1 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... \n",
361
+ "2 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... \n",
362
+ "3 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... \n",
363
+ "4 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMSLPGRWKPKM... \n",
364
+ "\n",
365
+ " smiles affinity_uM \n",
366
+ "0 COc1cc2c(Nc3ccc(Br)cc3F)ncnc2cc1OCC1CCN(C)CC1 0.00024 \n",
367
+ "1 O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(C\\C=C\\c2cn... 0.00025 \n",
368
+ "2 O[C@@H]1[C@@H](O)[C@@H](Cc2ccccc2)N(CC2CC2)C(=... 0.00041 \n",
369
+ "3 OCCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@... 0.00080 \n",
370
+ "4 OCCCCCN1[C@H](Cc2ccccc2)[C@H](O)[C@@H](O)[C@@H... 0.00099 "
371
+ ]
372
+ },
373
+ "execution_count": 89,
374
+ "metadata": {},
375
+ "output_type": "execute_result"
376
+ }
377
+ ],
378
+ "source": [
379
+ "df_bindingdb.head()"
380
+ ]
381
+ },
382
+ {
383
+ "cell_type": "code",
384
+ "execution_count": 93,
385
+ "id": "d7bfee2a-c4e6-48c9-b0c6-52f6a69c7453",
386
+ "metadata": {},
387
+ "outputs": [],
388
+ "source": [
389
+ "df_moad = pd.read_parquet('data/moad.parquet')\n",
390
+ "df_moad = df_moad[['seq','smiles','affinity_uM']]"
391
+ ]
392
+ },
393
+ {
394
+ "cell_type": "code",
395
+ "execution_count": 94,
396
+ "id": "25553199-1715-40fb-9260-427bdd6c3706",
397
+ "metadata": {},
398
+ "outputs": [
399
+ {
400
+ "data": {
401
+ "text/html": [
402
+ "<div>\n",
403
+ "<style scoped>\n",
404
+ " .dataframe tbody tr th:only-of-type {\n",
405
+ " vertical-align: middle;\n",
406
+ " }\n",
407
+ "\n",
408
+ " .dataframe tbody tr th {\n",
409
+ " vertical-align: top;\n",
410
+ " }\n",
411
+ "\n",
412
+ " .dataframe thead th {\n",
413
+ " text-align: right;\n",
414
+ " }\n",
415
+ "</style>\n",
416
+ "<table border=\"1\" class=\"dataframe\">\n",
417
+ " <thead>\n",
418
+ " <tr style=\"text-align: right;\">\n",
419
+ " <th></th>\n",
420
+ " <th>seq</th>\n",
421
+ " <th>smiles</th>\n",
422
+ " <th>affinity_uM</th>\n",
423
+ " </tr>\n",
424
+ " </thead>\n",
425
+ " <tbody>\n",
426
+ " <tr>\n",
427
+ " <th>0</th>\n",
428
+ " <td>NYIVPGEYRVAEGEIEINAGREKTTIRVSNTGDRPIQVGSHIHFVE...</td>\n",
429
+ " <td>NP(=O)(N)O</td>\n",
430
+ " <td>0.000620</td>\n",
431
+ " </tr>\n",
432
+ " <tr>\n",
433
+ " <th>2</th>\n",
434
+ " <td>NYIVPGEYRVAEGEIEINAGREKTTIRVSNTGDRPIQVGSHIHFVE...</td>\n",
435
+ " <td>CC(=O)NO</td>\n",
436
+ " <td>2.600000</td>\n",
437
+ " </tr>\n",
438
+ " <tr>\n",
439
+ " <th>7</th>\n",
440
+ " <td>MEGMRRPTPTVYVGRVPIGGAHPIAVQSMTNTPTRDVEATTAQVLE...</td>\n",
441
+ " <td>C#CCCOP(=O)(O)OP(=O)(O)O</td>\n",
442
+ " <td>0.580000</td>\n",
443
+ " </tr>\n",
444
+ " <tr>\n",
445
+ " <th>16</th>\n",
446
+ " <td>MEGMRRPTPTVYVGRVPIGGAHPIAVQSMTNTPTRDVEATTAQVLE...</td>\n",
447
+ " <td>C#CCOP(=O)(O)OP(=O)(O)O</td>\n",
448
+ " <td>0.770000</td>\n",
449
+ " </tr>\n",
450
+ " <tr>\n",
451
+ " <th>17</th>\n",
452
+ " <td>MTDMSIKFELIDVPIPQGTNVIIGQAHFIKTVEDLYEALVTSVPGV...</td>\n",
453
+ " <td>c1nc(c2c(n1)n(cn2)[C@H]3[C@@H]([C@@H]([C@H](O3...</td>\n",
454
+ " <td>15.000000</td>\n",
455
+ " </tr>\n",
456
+ " <tr>\n",
457
+ " <th>...</th>\n",
458
+ " <td>...</td>\n",
459
+ " <td>...</td>\n",
460
+ " <td>...</td>\n",
461
+ " </tr>\n",
462
+ " <tr>\n",
463
+ " <th>51900</th>\n",
464
+ " <td>MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG...</td>\n",
465
+ " <td>None</td>\n",
466
+ " <td>127.226463</td>\n",
467
+ " </tr>\n",
468
+ " <tr>\n",
469
+ " <th>51901</th>\n",
470
+ " <td>MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG...</td>\n",
471
+ " <td>None</td>\n",
472
+ " <td>127.226463</td>\n",
473
+ " </tr>\n",
474
+ " <tr>\n",
475
+ " <th>51902</th>\n",
476
+ " <td>MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG...</td>\n",
477
+ " <td>None</td>\n",
478
+ " <td>169.204738</td>\n",
479
+ " </tr>\n",
480
+ " <tr>\n",
481
+ " <th>51903</th>\n",
482
+ " <td>MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG...</td>\n",
483
+ " <td>None</td>\n",
484
+ " <td>169.204738</td>\n",
485
+ " </tr>\n",
486
+ " <tr>\n",
487
+ " <th>51904</th>\n",
488
+ " <td>MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG...</td>\n",
489
+ " <td>None</td>\n",
490
+ " <td>169.204738</td>\n",
491
+ " </tr>\n",
492
+ " </tbody>\n",
493
+ "</table>\n",
494
+ "<p>25425 rows × 3 columns</p>\n",
495
+ "</div>"
496
+ ],
497
+ "text/plain": [
498
+ " seq \\\n",
499
+ "0 NYIVPGEYRVAEGEIEINAGREKTTIRVSNTGDRPIQVGSHIHFVE... \n",
500
+ "2 NYIVPGEYRVAEGEIEINAGREKTTIRVSNTGDRPIQVGSHIHFVE... \n",
501
+ "7 MEGMRRPTPTVYVGRVPIGGAHPIAVQSMTNTPTRDVEATTAQVLE... \n",
502
+ "16 MEGMRRPTPTVYVGRVPIGGAHPIAVQSMTNTPTRDVEATTAQVLE... \n",
503
+ "17 MTDMSIKFELIDVPIPQGTNVIIGQAHFIKTVEDLYEALVTSVPGV... \n",
504
+ "... ... \n",
505
+ "51900 MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG... \n",
506
+ "51901 MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG... \n",
507
+ "51902 MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG... \n",
508
+ "51903 MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG... \n",
509
+ "51904 MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG... \n",
510
+ "\n",
511
+ " smiles affinity_uM \n",
512
+ "0 NP(=O)(N)O 0.000620 \n",
513
+ "2 CC(=O)NO 2.600000 \n",
514
+ "7 C#CCCOP(=O)(O)OP(=O)(O)O 0.580000 \n",
515
+ "16 C#CCOP(=O)(O)OP(=O)(O)O 0.770000 \n",
516
+ "17 c1nc(c2c(n1)n(cn2)[C@H]3[C@@H]([C@@H]([C@H](O3... 15.000000 \n",
517
+ "... ... ... \n",
518
+ "51900 None 127.226463 \n",
519
+ "51901 None 127.226463 \n",
520
+ "51902 None 169.204738 \n",
521
+ "51903 None 169.204738 \n",
522
+ "51904 None 169.204738 \n",
523
+ "\n",
524
+ "[25425 rows x 3 columns]"
525
+ ]
526
+ },
527
+ "execution_count": 94,
528
+ "metadata": {},
529
+ "output_type": "execute_result"
530
+ }
531
+ ],
532
+ "source": [
533
+ "df_moad"
534
+ ]
535
+ },
536
+ {
537
+ "cell_type": "code",
538
+ "execution_count": 97,
539
+ "id": "b2c936bc-cdc8-4bc1-b92d-f8755fd65f0a",
540
+ "metadata": {},
541
+ "outputs": [],
542
+ "source": [
543
+ "df_biolip = pd.read_parquet('data/biolip.parquet')\n",
544
+ "df_biolip = df_biolip[['seq','smiles','affinity_uM']]"
545
+ ]
546
+ },
547
+ {
548
+ "cell_type": "code",
549
+ "execution_count": 98,
550
+ "id": "cee93018-601d-458b-af44-bd978da7a2bc",
551
+ "metadata": {},
552
+ "outputs": [
553
+ {
554
+ "data": {
555
+ "text/html": [
556
+ "<div>\n",
557
+ "<style scoped>\n",
558
+ " .dataframe tbody tr th:only-of-type {\n",
559
+ " vertical-align: middle;\n",
560
+ " }\n",
561
+ "\n",
562
+ " .dataframe tbody tr th {\n",
563
+ " vertical-align: top;\n",
564
+ " }\n",
565
+ "\n",
566
+ " .dataframe thead th {\n",
567
+ " text-align: right;\n",
568
+ " }\n",
569
+ "</style>\n",
570
+ "<table border=\"1\" class=\"dataframe\">\n",
571
+ " <thead>\n",
572
+ " <tr style=\"text-align: right;\">\n",
573
+ " <th></th>\n",
574
+ " <th>seq</th>\n",
575
+ " <th>smiles</th>\n",
576
+ " <th>affinity_uM</th>\n",
577
+ " </tr>\n",
578
+ " </thead>\n",
579
+ " <tbody>\n",
580
+ " <tr>\n",
581
+ " <th>38</th>\n",
582
+ " <td>PYTVVYFPVRGRCAALRMLLADQGQSWKEEVVTVETWQEGSLKASC...</td>\n",
583
+ " <td>CC[C@H](C(=O)c1ccc(c(c1Cl)Cl)OCC(=O)O)C</td>\n",
584
+ " <td>1.500</td>\n",
585
+ " </tr>\n",
586
+ " <tr>\n",
587
+ " <th>43</th>\n",
588
+ " <td>MPPYTVVYFPVRGRCAALRMLLADQGQSWKEEVVTVETWQEGSLKA...</td>\n",
589
+ " <td>OC(=O)c1cc(/N=N/c2ccc(cc2)S(=O)(=O)Nc2ccccn2)c...</td>\n",
590
+ " <td>24.000</td>\n",
591
+ " </tr>\n",
592
+ " <tr>\n",
593
+ " <th>53</th>\n",
594
+ " <td>EKKSINECDLKGKKVLIRVDFNVPVKNGKITNDYRIRSALPTLKKV...</td>\n",
595
+ " <td>O[C@@H]1[C@@H](CO[P@](=O)(O[P@@](=O)(C(CCCC(P(...</td>\n",
596
+ " <td>NaN</td>\n",
597
+ " </tr>\n",
598
+ " <tr>\n",
599
+ " <th>54</th>\n",
600
+ " <td>MPPYTVVYFPVRGRCAALRMLLADQGQSWKEEVVTVETWQEGSLKA...</td>\n",
601
+ " <td>CCCCCCSC[C@@H](C(=O)NCC(=O)O)NC(=O)CC[C@@H](C(...</td>\n",
602
+ " <td>10.000</td>\n",
603
+ " </tr>\n",
604
+ " <tr>\n",
605
+ " <th>55</th>\n",
606
+ " <td>MNIFEMLRIDEGLRLKIYKDTEGYYTIGIGHLLTKSPSLNAAKSEL...</td>\n",
607
+ " <td>c1ccccc1</td>\n",
608
+ " <td>175.000</td>\n",
609
+ " </tr>\n",
610
+ " <tr>\n",
611
+ " <th>...</th>\n",
612
+ " <td>...</td>\n",
613
+ " <td>...</td>\n",
614
+ " <td>...</td>\n",
615
+ " </tr>\n",
616
+ " <tr>\n",
617
+ " <th>105118</th>\n",
618
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMNLPGRWKPKM...</td>\n",
619
+ " <td>O[C@@H]([C@H](Cc1ccccc1)NC(=O)[C@H](C(C)C)NC(=...</td>\n",
620
+ " <td>NaN</td>\n",
621
+ " </tr>\n",
622
+ " <tr>\n",
623
+ " <th>105119</th>\n",
624
+ " <td>PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMNLPGRWKPKM...</td>\n",
625
+ " <td>O[C@@H]([C@H](Cc1ccccc1)NC(=O)[C@H](C(C)C)NC(=...</td>\n",
626
+ " <td>NaN</td>\n",
627
+ " </tr>\n",
628
+ " <tr>\n",
629
+ " <th>105124</th>\n",
630
+ " <td>SKVVVPAQGKKITLQNGKLNVPENPIIPYIEGDGIGVDVTPAMLKV...</td>\n",
631
+ " <td>O[C@@H]1[C@@H](COP(=O)(O)O)O[C@H]([C@@H]1OP(=O...</td>\n",
632
+ " <td>125.000</td>\n",
633
+ " </tr>\n",
634
+ " <tr>\n",
635
+ " <th>105133</th>\n",
636
+ " <td>ANIVGGIEYSINNASLCSVGFSVTRGATKGFVTAGHCGTVNATARI...</td>\n",
637
+ " <td>CC(C[C@@H](B(O)O)NC(=O)[C@@H]1CCCN1C(=O)[C@@H]...</td>\n",
638
+ " <td>NaN</td>\n",
639
+ " </tr>\n",
640
+ " <tr>\n",
641
+ " <th>105138</th>\n",
642
+ " <td>KFPRVKNWELGSITYDTLCAQSQQDGPCTPRRCLGSLVLPRKLQTR...</td>\n",
643
+ " <td>CC[Se]C(=N)N</td>\n",
644
+ " <td>0.039</td>\n",
645
+ " </tr>\n",
646
+ " </tbody>\n",
647
+ "</table>\n",
648
+ "<p>13645 rows × 3 columns</p>\n",
649
+ "</div>"
650
+ ],
651
+ "text/plain": [
652
+ " seq \\\n",
653
+ "38 PYTVVYFPVRGRCAALRMLLADQGQSWKEEVVTVETWQEGSLKASC... \n",
654
+ "43 MPPYTVVYFPVRGRCAALRMLLADQGQSWKEEVVTVETWQEGSLKA... \n",
655
+ "53 EKKSINECDLKGKKVLIRVDFNVPVKNGKITNDYRIRSALPTLKKV... \n",
656
+ "54 MPPYTVVYFPVRGRCAALRMLLADQGQSWKEEVVTVETWQEGSLKA... \n",
657
+ "55 MNIFEMLRIDEGLRLKIYKDTEGYYTIGIGHLLTKSPSLNAAKSEL... \n",
658
+ "... ... \n",
659
+ "105118 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMNLPGRWKPKM... \n",
660
+ "105119 PQITLWQRPLVTIKIGGQLKEALLDTGADDTVLEEMNLPGRWKPKM... \n",
661
+ "105124 SKVVVPAQGKKITLQNGKLNVPENPIIPYIEGDGIGVDVTPAMLKV... \n",
662
+ "105133 ANIVGGIEYSINNASLCSVGFSVTRGATKGFVTAGHCGTVNATARI... \n",
663
+ "105138 KFPRVKNWELGSITYDTLCAQSQQDGPCTPRRCLGSLVLPRKLQTR... \n",
664
+ "\n",
665
+ " smiles affinity_uM \n",
666
+ "38 CC[C@H](C(=O)c1ccc(c(c1Cl)Cl)OCC(=O)O)C 1.500 \n",
667
+ "43 OC(=O)c1cc(/N=N/c2ccc(cc2)S(=O)(=O)Nc2ccccn2)c... 24.000 \n",
668
+ "53 O[C@@H]1[C@@H](CO[P@](=O)(O[P@@](=O)(C(CCCC(P(... NaN \n",
669
+ "54 CCCCCCSC[C@@H](C(=O)NCC(=O)O)NC(=O)CC[C@@H](C(... 10.000 \n",
670
+ "55 c1ccccc1 175.000 \n",
671
+ "... ... ... \n",
672
+ "105118 O[C@@H]([C@H](Cc1ccccc1)NC(=O)[C@H](C(C)C)NC(=... NaN \n",
673
+ "105119 O[C@@H]([C@H](Cc1ccccc1)NC(=O)[C@H](C(C)C)NC(=... NaN \n",
674
+ "105124 O[C@@H]1[C@@H](COP(=O)(O)O)O[C@H]([C@@H]1OP(=O... 125.000 \n",
675
+ "105133 CC(C[C@@H](B(O)O)NC(=O)[C@@H]1CCCN1C(=O)[C@@H]... NaN \n",
676
+ "105138 CC[Se]C(=N)N 0.039 \n",
677
+ "\n",
678
+ "[13645 rows x 3 columns]"
679
+ ]
680
+ },
681
+ "execution_count": 98,
682
+ "metadata": {},
683
+ "output_type": "execute_result"
684
+ }
685
+ ],
686
+ "source": [
687
+ "df_biolip"
688
+ ]
689
+ },
690
+ {
691
+ "cell_type": "code",
692
+ "execution_count": 134,
693
+ "id": "195f92db-fe06-4d03-8500-8d6c310a3347",
694
+ "metadata": {},
695
+ "outputs": [],
696
+ "source": [
697
+ "df_all = pd.concat([df_pdbbind,df_bindingdb,df_moad,df_biolip]).reset_index()"
698
+ ]
699
+ },
700
+ {
701
+ "cell_type": "code",
702
+ "execution_count": 135,
703
+ "id": "d25c1e24-6566-4944-a0b4-944b3c8dbc6f",
704
+ "metadata": {},
705
+ "outputs": [
706
+ {
707
+ "data": {
708
+ "text/plain": [
709
+ "2446422"
710
+ ]
711
+ },
712
+ "execution_count": 135,
713
+ "metadata": {},
714
+ "output_type": "execute_result"
715
+ }
716
+ ],
717
+ "source": [
718
+ "len(df_all)"
719
+ ]
720
+ },
721
+ {
722
+ "cell_type": "code",
723
+ "execution_count": 105,
724
+ "id": "c8287da2-cfdf-4d89-b175-f4c6b38ff8ac",
725
+ "metadata": {},
726
+ "outputs": [
727
+ {
728
+ "name": "stdout",
729
+ "output_type": "stream",
730
+ "text": [
731
+ "INFO: Pandarallel will run on 32 workers.\n",
732
+ "INFO: Pandarallel will use Memory file system to transfer data between the main process and workers.\n"
733
+ ]
734
+ }
735
+ ],
736
+ "source": [
737
+ "from pandarallel import pandarallel\n",
738
+ "pandarallel.initialize()"
739
+ ]
740
+ },
741
+ {
742
+ "cell_type": "code",
743
+ "execution_count": null,
744
+ "id": "de5ffc4a-afb7-4a26-8d57-509c2278d750",
745
+ "metadata": {},
746
+ "outputs": [],
747
+ "source": [
748
+ "df_all['maccs'] = df_all['smiles'].parallel_apply(get_maccs)"
749
+ ]
750
+ },
751
+ {
752
+ "cell_type": "code",
753
+ "execution_count": 108,
754
+ "id": "59a6706d-dab9-4ee0-8ef6-33537a3622a4",
755
+ "metadata": {},
756
+ "outputs": [],
757
+ "source": [
758
+ "df_all.to_parquet('data/all_maccs.parquet')"
759
+ ]
760
+ },
761
+ {
762
+ "cell_type": "code",
763
+ "execution_count": 6,
764
+ "id": "4ccf2ee5-d369-4c0e-bb91-792765d661bf",
765
+ "metadata": {},
766
+ "outputs": [],
767
+ "source": [
768
+ "import numpy as np"
769
+ ]
770
+ },
771
+ {
772
+ "cell_type": "code",
773
+ "execution_count": 14,
774
+ "id": "8a4bbb18-e62f-4774-ac6b-8a1be68204c1",
775
+ "metadata": {},
776
+ "outputs": [],
777
+ "source": [
778
+ "df_all = pd.read_parquet('data/all_maccs.parquet')\n",
779
+ "df_all = df_all.dropna().reset_index(drop=True)"
780
+ ]
781
+ },
782
+ {
783
+ "cell_type": "code",
784
+ "execution_count": 15,
785
+ "id": "d210fe56-a7eb-4adc-a77a-14c0c6d0034e",
786
+ "metadata": {},
787
+ "outputs": [
788
+ {
789
+ "data": {
790
+ "text/plain": [
791
+ "2430135"
792
+ ]
793
+ },
794
+ "execution_count": 15,
795
+ "metadata": {},
796
+ "output_type": "execute_result"
797
+ }
798
+ ],
799
+ "source": [
800
+ "len(df_all)"
801
+ ]
802
+ },
803
+ {
804
+ "cell_type": "code",
805
+ "execution_count": 16,
806
+ "id": "d12b365d-98bd-4b61-b836-1a08d2e55418",
807
+ "metadata": {},
808
+ "outputs": [],
809
+ "source": [
810
+ "maccs = df_all['maccs'].to_numpy()\n",
811
+ "#df_reindex[df_reindex.duplicated(keep='first')].reset_index()"
812
+ ]
813
+ },
814
+ {
815
+ "cell_type": "code",
816
+ "execution_count": 17,
817
+ "id": "80c15210-1af3-436e-970b-f81fc596fb41",
818
+ "metadata": {},
819
+ "outputs": [],
820
+ "source": [
821
+ "df_maccs = pd.DataFrame(np.vstack(maccs))"
822
+ ]
823
+ },
824
+ {
825
+ "cell_type": "code",
826
+ "execution_count": 18,
827
+ "id": "30c314b8-8fe7-48ae-a2b8-149de1471b0c",
828
+ "metadata": {},
829
+ "outputs": [
830
+ {
831
+ "data": {
832
+ "text/plain": [
833
+ "0 int64\n",
834
+ "1 int64\n",
835
+ "2 int64\n",
836
+ "3 int64\n",
837
+ "4 int64\n",
838
+ "5 int64\n",
839
+ "dtype: object"
840
+ ]
841
+ },
842
+ "execution_count": 18,
843
+ "metadata": {},
844
+ "output_type": "execute_result"
845
+ }
846
+ ],
847
+ "source": [
848
+ "df_maccs.dtypes"
849
+ ]
850
+ },
851
+ {
852
+ "cell_type": "code",
853
+ "execution_count": 19,
854
+ "id": "70a0a820-4d0c-4472-af96-9c301c0ab204",
855
+ "metadata": {},
856
+ "outputs": [],
857
+ "source": [
858
+ "df_expand = pd.concat([df_all[['seq','smiles','affinity_uM']],df_maccs],axis=1)"
859
+ ]
860
+ },
861
+ {
862
+ "cell_type": "code",
863
+ "execution_count": 21,
864
+ "id": "13d092fa-5625-40d0-b7ec-e3405ea20279",
865
+ "metadata": {},
866
+ "outputs": [
867
+ {
868
+ "data": {
869
+ "text/html": [
870
+ "<div>\n",
871
+ "<style scoped>\n",
872
+ " .dataframe tbody tr th:only-of-type {\n",
873
+ " vertical-align: middle;\n",
874
+ " }\n",
875
+ "\n",
876
+ " .dataframe tbody tr th {\n",
877
+ " vertical-align: top;\n",
878
+ " }\n",
879
+ "\n",
880
+ " .dataframe thead th {\n",
881
+ " text-align: right;\n",
882
+ " }\n",
883
+ "</style>\n",
884
+ "<table border=\"1\" class=\"dataframe\">\n",
885
+ " <thead>\n",
886
+ " <tr style=\"text-align: right;\">\n",
887
+ " <th></th>\n",
888
+ " <th>seq</th>\n",
889
+ " <th>smiles</th>\n",
890
+ " <th>affinity_uM</th>\n",
891
+ " <th>0</th>\n",
892
+ " <th>1</th>\n",
893
+ " <th>2</th>\n",
894
+ " <th>3</th>\n",
895
+ " <th>4</th>\n",
896
+ " <th>5</th>\n",
897
+ " </tr>\n",
898
+ " </thead>\n",
899
+ " <tbody>\n",
900
+ " <tr>\n",
901
+ " <th>0</th>\n",
902
+ " <td>APQTITELCSEYRNTQIYTINDKILSYTESMAGKREMVIITFKSGE...</td>\n",
903
+ " <td>OC[C@H]1O[C@H](Oc2cccc(c2)N(=O)=O)[C@@H]([C@H]...</td>\n",
904
+ " <td>500.000</td>\n",
905
+ " <td>2147483648</td>\n",
906
+ " <td>3242590208</td>\n",
907
+ " <td>1914732547</td>\n",
908
+ " <td>994116706</td>\n",
909
+ " <td>3748288829</td>\n",
910
+ " <td>124</td>\n",
911
+ " </tr>\n",
912
+ " <tr>\n",
913
+ " <th>1</th>\n",
914
+ " <td>VETFAFQAEIAQLMSLIINTFYSNKEIFLRELISNSSDALDKIRYE...</td>\n",
915
+ " <td>COc1ccc(cc1)c1c(onc1c1cc(C(C)C)c(cc1O)O)NC(=O)...</td>\n",
916
+ " <td>0.023</td>\n",
917
+ " <td>131072</td>\n",
918
+ " <td>1109655552</td>\n",
919
+ " <td>2123376961</td>\n",
920
+ " <td>3477340882</td>\n",
921
+ " <td>2951175957</td>\n",
922
+ " <td>252</td>\n",
923
+ " </tr>\n",
924
+ " <tr>\n",
925
+ " <th>2</th>\n",
926
+ " <td>GMRVYLGADHAGYELKQRIIEHLKQTGHEPIDCGALRYDADDDYPA...</td>\n",
927
+ " <td>O[C@H]1O[C@H](CO[P](=O)(=O)=O)[C@H]([C@H]([C@H...</td>\n",
928
+ " <td>6300.000</td>\n",
929
+ " <td>67108864</td>\n",
930
+ " <td>1082523648</td>\n",
931
+ " <td>1879080960</td>\n",
932
+ " <td>461382690</td>\n",
933
+ " <td>3576355128</td>\n",
934
+ " <td>28</td>\n",
935
+ " </tr>\n",
936
+ " <tr>\n",
937
+ " <th>3</th>\n",
938
+ " <td>SMENFQKVEKIGEGTYGVVYKARNKLTGEVVALKKIRLDTETEGVP...</td>\n",
939
+ " <td>OCC[C@@H]1CCCCN1c1cc(NCC2=CC=CN(C2)O)n2c(n1)c(...</td>\n",
940
+ " <td>0.210</td>\n",
941
+ " <td>2147484672</td>\n",
942
+ " <td>36176898</td>\n",
943
+ " <td>850664773</td>\n",
944
+ " <td>3978479102</td>\n",
945
+ " <td>1599828989</td>\n",
946
+ " <td>252</td>\n",
947
+ " </tr>\n",
948
+ " <tr>\n",
949
+ " <th>4</th>\n",
950
+ " <td>EFSEWFHNILEEAEIIDQRYPVKGMHVWMPHGFMIRKNTLKILRRI...</td>\n",
951
+ " <td>O[C@@H]1[C@@H](COS(=O)(=O)NC(=O)[C@@H]2CCC[NH2...</td>\n",
952
+ " <td>0.050</td>\n",
953
+ " <td>0</td>\n",
954
+ " <td>1858306115</td>\n",
955
+ " <td>4223456596</td>\n",
956
+ " <td>4018595822</td>\n",
957
+ " <td>4282121085</td>\n",
958
+ " <td>124</td>\n",
959
+ " </tr>\n",
960
+ " <tr>\n",
961
+ " <th>...</th>\n",
962
+ " <td>...</td>\n",
963
+ " <td>...</td>\n",
964
+ " <td>...</td>\n",
965
+ " <td>...</td>\n",
966
+ " <td>...</td>\n",
967
+ " <td>...</td>\n",
968
+ " <td>...</td>\n",
969
+ " <td>...</td>\n",
970
+ " <td>...</td>\n",
971
+ " </tr>\n",
972
+ " <tr>\n",
973
+ " <th>2430130</th>\n",
974
+ " <td>IVEGSDAEIGMSPWQVMLFRKSPQELLCGASLISDRWVLTAAHCLL...</td>\n",
975
+ " <td>O=C[C@@H](NC(=O)[C@H](Cc1ccc(cc1)OS(O)(O)O)NC(...</td>\n",
976
+ " <td>8.000</td>\n",
977
+ " <td>0</td>\n",
978
+ " <td>612865025</td>\n",
979
+ " <td>3107729684</td>\n",
980
+ " <td>2146870234</td>\n",
981
+ " <td>4286578680</td>\n",
982
+ " <td>252</td>\n",
983
+ " </tr>\n",
984
+ " <tr>\n",
985
+ " <th>2430131</th>\n",
986
+ " <td>IVEGSDAEIGMSPWQVMLFRKSPQELLCGASLISDRWVLTAAHCLL...</td>\n",
987
+ " <td>CC(C[C@@H](C(=O)N1C=CC[C@H]1C(=O)N)NC(=O)[C@@H...</td>\n",
988
+ " <td>8.000</td>\n",
989
+ " <td>0</td>\n",
990
+ " <td>136194</td>\n",
991
+ " <td>1025390336</td>\n",
992
+ " <td>1612680088</td>\n",
993
+ " <td>2071973584</td>\n",
994
+ " <td>252</td>\n",
995
+ " </tr>\n",
996
+ " <tr>\n",
997
+ " <th>2430132</th>\n",
998
+ " <td>RWEQTHLTYRIENYTPDLPRADVDHAIEKAFQLWSNVTPLTFTKVS...</td>\n",
999
+ " <td>ONC(=O)CC1(CCOCC1)S(=O)(=O)c1ccc(cc1)Oc1ccccc1</td>\n",
1000
+ " <td>0.023</td>\n",
1001
+ " <td>2147483648</td>\n",
1002
+ " <td>2081488896</td>\n",
1003
+ " <td>3124936893</td>\n",
1004
+ " <td>264668962</td>\n",
1005
+ " <td>4286183928</td>\n",
1006
+ " <td>124</td>\n",
1007
+ " </tr>\n",
1008
+ " <tr>\n",
1009
+ " <th>2430133</th>\n",
1010
+ " <td>SKVVVPAQGKKITLQNGKLNVPENPIIPYIEGDGIGVDVTPAMLKV...</td>\n",
1011
+ " <td>O[C@@H]1[C@@H](COP(=O)(O)O)O[C@H]([C@@H]1OP(=O...</td>\n",
1012
+ " <td>125.000</td>\n",
1013
+ " <td>67108864</td>\n",
1014
+ " <td>1115688962</td>\n",
1015
+ " <td>1771869508</td>\n",
1016
+ " <td>4018431718</td>\n",
1017
+ " <td>3744193341</td>\n",
1018
+ " <td>124</td>\n",
1019
+ " </tr>\n",
1020
+ " <tr>\n",
1021
+ " <th>2430134</th>\n",
1022
+ " <td>KFPRVKNWELGSITYDTLCAQSQQDGPCTPRRCLGSLVLPRKLQTR...</td>\n",
1023
+ " <td>CC[Se]C(=N)N</td>\n",
1024
+ " <td>0.039</td>\n",
1025
+ " <td>16</td>\n",
1026
+ " <td>6144</td>\n",
1027
+ " <td>537396736</td>\n",
1028
+ " <td>2170880</td>\n",
1029
+ " <td>1510015504</td>\n",
1030
+ " <td>192</td>\n",
1031
+ " </tr>\n",
1032
+ " </tbody>\n",
1033
+ "</table>\n",
1034
+ "<p>2430135 rows × 9 columns</p>\n",
1035
+ "</div>"
1036
+ ],
1037
+ "text/plain": [
1038
+ " seq \\\n",
1039
+ "0 APQTITELCSEYRNTQIYTINDKILSYTESMAGKREMVIITFKSGE... \n",
1040
+ "1 VETFAFQAEIAQLMSLIINTFYSNKEIFLRELISNSSDALDKIRYE... \n",
1041
+ "2 GMRVYLGADHAGYELKQRIIEHLKQTGHEPIDCGALRYDADDDYPA... \n",
1042
+ "3 SMENFQKVEKIGEGTYGVVYKARNKLTGEVVALKKIRLDTETEGVP... \n",
1043
+ "4 EFSEWFHNILEEAEIIDQRYPVKGMHVWMPHGFMIRKNTLKILRRI... \n",
1044
+ "... ... \n",
1045
+ "2430130 IVEGSDAEIGMSPWQVMLFRKSPQELLCGASLISDRWVLTAAHCLL... \n",
1046
+ "2430131 IVEGSDAEIGMSPWQVMLFRKSPQELLCGASLISDRWVLTAAHCLL... \n",
1047
+ "2430132 RWEQTHLTYRIENYTPDLPRADVDHAIEKAFQLWSNVTPLTFTKVS... \n",
1048
+ "2430133 SKVVVPAQGKKITLQNGKLNVPENPIIPYIEGDGIGVDVTPAMLKV... \n",
1049
+ "2430134 KFPRVKNWELGSITYDTLCAQSQQDGPCTPRRCLGSLVLPRKLQTR... \n",
1050
+ "\n",
1051
+ " smiles affinity_uM \\\n",
1052
+ "0 OC[C@H]1O[C@H](Oc2cccc(c2)N(=O)=O)[C@@H]([C@H]... 500.000 \n",
1053
+ "1 COc1ccc(cc1)c1c(onc1c1cc(C(C)C)c(cc1O)O)NC(=O)... 0.023 \n",
1054
+ "2 O[C@H]1O[C@H](CO[P](=O)(=O)=O)[C@H]([C@H]([C@H... 6300.000 \n",
1055
+ "3 OCC[C@@H]1CCCCN1c1cc(NCC2=CC=CN(C2)O)n2c(n1)c(... 0.210 \n",
1056
+ "4 O[C@@H]1[C@@H](COS(=O)(=O)NC(=O)[C@@H]2CCC[NH2... 0.050 \n",
1057
+ "... ... ... \n",
1058
+ "2430130 O=C[C@@H](NC(=O)[C@H](Cc1ccc(cc1)OS(O)(O)O)NC(... 8.000 \n",
1059
+ "2430131 CC(C[C@@H](C(=O)N1C=CC[C@H]1C(=O)N)NC(=O)[C@@H... 8.000 \n",
1060
+ "2430132 ONC(=O)CC1(CCOCC1)S(=O)(=O)c1ccc(cc1)Oc1ccccc1 0.023 \n",
1061
+ "2430133 O[C@@H]1[C@@H](COP(=O)(O)O)O[C@H]([C@@H]1OP(=O... 125.000 \n",
1062
+ "2430134 CC[Se]C(=N)N 0.039 \n",
1063
+ "\n",
1064
+ " 0 1 2 3 4 5 \n",
1065
+ "0 2147483648 3242590208 1914732547 994116706 3748288829 124 \n",
1066
+ "1 131072 1109655552 2123376961 3477340882 2951175957 252 \n",
1067
+ "2 67108864 1082523648 1879080960 461382690 3576355128 28 \n",
1068
+ "3 2147484672 36176898 850664773 3978479102 1599828989 252 \n",
1069
+ "4 0 1858306115 4223456596 4018595822 4282121085 124 \n",
1070
+ "... ... ... ... ... ... ... \n",
1071
+ "2430130 0 612865025 3107729684 2146870234 4286578680 252 \n",
1072
+ "2430131 0 136194 1025390336 1612680088 2071973584 252 \n",
1073
+ "2430132 2147483648 2081488896 3124936893 264668962 4286183928 124 \n",
1074
+ "2430133 67108864 1115688962 1771869508 4018431718 3744193341 124 \n",
1075
+ "2430134 16 6144 537396736 2170880 1510015504 192 \n",
1076
+ "\n",
1077
+ "[2430135 rows x 9 columns]"
1078
+ ]
1079
+ },
1080
+ "execution_count": 21,
1081
+ "metadata": {},
1082
+ "output_type": "execute_result"
1083
+ }
1084
+ ],
1085
+ "source": [
1086
+ "df_expand"
1087
+ ]
1088
+ },
1089
+ {
1090
+ "cell_type": "code",
1091
+ "execution_count": 22,
1092
+ "id": "30f7fff7-3cfe-41c8-97c9-666f3e256222",
1093
+ "metadata": {},
1094
+ "outputs": [
1095
+ {
1096
+ "data": {
1097
+ "text/plain": [
1098
+ "Index(['seq', 'smiles', 'affinity_uM', 0, 1, 2, 3, 4, 5], dtype='object')"
1099
+ ]
1100
+ },
1101
+ "execution_count": 22,
1102
+ "metadata": {},
1103
+ "output_type": "execute_result"
1104
+ }
1105
+ ],
1106
+ "source": [
1107
+ "df_expand.columns"
1108
+ ]
1109
+ },
1110
+ {
1111
+ "cell_type": "code",
1112
+ "execution_count": 23,
1113
+ "id": "16d2b26e-984f-4c71-af19-a3e711ed9ca2",
1114
+ "metadata": {},
1115
+ "outputs": [],
1116
+ "source": [
1117
+ "df_reindex = df_expand.set_index([0,1,2,3,4,5,'seq'])"
1118
+ ]
1119
+ },
1120
+ {
1121
+ "cell_type": "code",
1122
+ "execution_count": 24,
1123
+ "id": "27fa2150-8152-444b-ba5b-24bea39fc098",
1124
+ "metadata": {},
1125
+ "outputs": [
1126
+ {
1127
+ "data": {
1128
+ "text/plain": [
1129
+ "Index(['smiles', 'affinity_uM'], dtype='object')"
1130
+ ]
1131
+ },
1132
+ "execution_count": 24,
1133
+ "metadata": {},
1134
+ "output_type": "execute_result"
1135
+ }
1136
+ ],
1137
+ "source": [
1138
+ "df_reindex.columns"
1139
+ ]
1140
+ },
1141
+ {
1142
+ "cell_type": "code",
1143
+ "execution_count": 67,
1144
+ "id": "89edacbc-52f3-4a76-90b0-95273f5e53b3",
1145
+ "metadata": {},
1146
+ "outputs": [],
1147
+ "source": [
1148
+ "df_nr = df_reindex[~df_reindex.duplicated(keep='first')].reset_index()\n",
1149
+ "df_nr = df_nr.drop(columns=[0,1,2,3,4,5])"
1150
+ ]
1151
+ },
1152
+ {
1153
+ "cell_type": "code",
1154
+ "execution_count": 68,
1155
+ "id": "6a704c5e-68a6-418f-bcad-8688a13ca1d6",
1156
+ "metadata": {},
1157
+ "outputs": [],
1158
+ "source": [
1159
+ "# final sanity checks"
1160
+ ]
1161
+ },
1162
+ {
1163
+ "cell_type": "code",
1164
+ "execution_count": 69,
1165
+ "id": "0cad3882-975d-4693-aad1-63ec26646bd0",
1166
+ "metadata": {},
1167
+ "outputs": [
1168
+ {
1169
+ "name": "stderr",
1170
+ "output_type": "stream",
1171
+ "text": [
1172
+ "/ccs/proj/stf006/glaser/conda-envs/bio/lib/python3.9/site-packages/pandas/core/arraylike.py:358: RuntimeWarning: divide by zero encountered in log\n",
1173
+ " result = getattr(ufunc, method)(*inputs, **kwargs)\n"
1174
+ ]
1175
+ }
1176
+ ],
1177
+ "source": [
1178
+ "df_nr['neg_log10_affinity_M'] = 6-np.log(df_nr['affinity_uM'])/np.log(10)"
1179
+ ]
1180
+ },
1181
+ {
1182
+ "cell_type": "code",
1183
+ "execution_count": 70,
1184
+ "id": "c200e29a-3f14-41f4-b620-ccce0eb0d5ce",
1185
+ "metadata": {},
1186
+ "outputs": [
1187
+ {
1188
+ "data": {
1189
+ "text/html": [
1190
+ "<div>\n",
1191
+ "<style scoped>\n",
1192
+ " .dataframe tbody tr th:only-of-type {\n",
1193
+ " vertical-align: middle;\n",
1194
+ " }\n",
1195
+ "\n",
1196
+ " .dataframe tbody tr th {\n",
1197
+ " vertical-align: top;\n",
1198
+ " }\n",
1199
+ "\n",
1200
+ " .dataframe thead th {\n",
1201
+ " text-align: right;\n",
1202
+ " }\n",
1203
+ "</style>\n",
1204
+ "<table border=\"1\" class=\"dataframe\">\n",
1205
+ " <thead>\n",
1206
+ " <tr style=\"text-align: right;\">\n",
1207
+ " <th></th>\n",
1208
+ " <th>seq</th>\n",
1209
+ " <th>smiles</th>\n",
1210
+ " <th>affinity_uM</th>\n",
1211
+ " <th>neg_log10_affinity_M</th>\n",
1212
+ " </tr>\n",
1213
+ " </thead>\n",
1214
+ " <tbody>\n",
1215
+ " <tr>\n",
1216
+ " <th>0</th>\n",
1217
+ " <td>APQTITELCSEYRNTQIYTINDKILSYTESMAGKREMVIITFKSGE...</td>\n",
1218
+ " <td>OC[C@H]1O[C@H](Oc2cccc(c2)N(=O)=O)[C@@H]([C@H]...</td>\n",
1219
+ " <td>500.000</td>\n",
1220
+ " <td>3.301030</td>\n",
1221
+ " </tr>\n",
1222
+ " <tr>\n",
1223
+ " <th>1</th>\n",
1224
+ " <td>VETFAFQAEIAQLMSLIINTFYSNKEIFLRELISNSSDALDKIRYE...</td>\n",
1225
+ " <td>COc1ccc(cc1)c1c(onc1c1cc(C(C)C)c(cc1O)O)NC(=O)...</td>\n",
1226
+ " <td>0.023</td>\n",
1227
+ " <td>7.638272</td>\n",
1228
+ " </tr>\n",
1229
+ " <tr>\n",
1230
+ " <th>2</th>\n",
1231
+ " <td>GMRVYLGADHAGYELKQRIIEHLKQTGHEPIDCGALRYDADDDYPA...</td>\n",
1232
+ " <td>O[C@H]1O[C@H](CO[P](=O)(=O)=O)[C@H]([C@H]([C@H...</td>\n",
1233
+ " <td>6300.000</td>\n",
1234
+ " <td>2.200659</td>\n",
1235
+ " </tr>\n",
1236
+ " <tr>\n",
1237
+ " <th>3</th>\n",
1238
+ " <td>SMENFQKVEKIGEGTYGVVYKARNKLTGEVVALKKIRLDTETEGVP...</td>\n",
1239
+ " <td>OCC[C@@H]1CCCCN1c1cc(NCC2=CC=CN(C2)O)n2c(n1)c(...</td>\n",
1240
+ " <td>0.210</td>\n",
1241
+ " <td>6.677781</td>\n",
1242
+ " </tr>\n",
1243
+ " <tr>\n",
1244
+ " <th>4</th>\n",
1245
+ " <td>EFSEWFHNILEEAEIIDQRYPVKGMHVWMPHGFMIRKNTLKILRRI...</td>\n",
1246
+ " <td>O[C@@H]1[C@@H](COS(=O)(=O)NC(=O)[C@@H]2CCC[NH2...</td>\n",
1247
+ " <td>0.050</td>\n",
1248
+ " <td>7.301030</td>\n",
1249
+ " </tr>\n",
1250
+ " <tr>\n",
1251
+ " <th>...</th>\n",
1252
+ " <td>...</td>\n",
1253
+ " <td>...</td>\n",
1254
+ " <td>...</td>\n",
1255
+ " <td>...</td>\n",
1256
+ " </tr>\n",
1257
+ " <tr>\n",
1258
+ " <th>1849400</th>\n",
1259
+ " <td>KQISVRGLAGVENVTELKKNFNRHLHFTLVKDRNVATPRDYYFALA...</td>\n",
1260
+ " <td>O[C@@H]1[C@H](O)[C@H](O[C@H]1n1cnc2c1ncnc2N)CO...</td>\n",
1261
+ " <td>250.000</td>\n",
1262
+ " <td>3.602060</td>\n",
1263
+ " </tr>\n",
1264
+ " <tr>\n",
1265
+ " <th>1849401</th>\n",
1266
+ " <td>IVEGSDAEIGMSPWQVMLFRKSPQELLCGASLISDRWVLTAAHCLL...</td>\n",
1267
+ " <td>O=C[C@@H](NC(=O)[C@H](Cc1ccc(cc1)OS(O)(O)O)NC(...</td>\n",
1268
+ " <td>8.000</td>\n",
1269
+ " <td>5.096910</td>\n",
1270
+ " </tr>\n",
1271
+ " <tr>\n",
1272
+ " <th>1849402</th>\n",
1273
+ " <td>IVEGSDAEIGMSPWQVMLFRKSPQELLCGASLISDRWVLTAAHCLL...</td>\n",
1274
+ " <td>CC(C[C@@H](C(=O)N1C=CC[C@H]1C(=O)N)NC(=O)[C@@H...</td>\n",
1275
+ " <td>8.000</td>\n",
1276
+ " <td>5.096910</td>\n",
1277
+ " </tr>\n",
1278
+ " <tr>\n",
1279
+ " <th>1849403</th>\n",
1280
+ " <td>SKVVVPAQGKKITLQNGKLNVPENPIIPYIEGDGIGVDVTPAMLKV...</td>\n",
1281
+ " <td>O[C@@H]1[C@@H](COP(=O)(O)O)O[C@H]([C@@H]1OP(=O...</td>\n",
1282
+ " <td>125.000</td>\n",
1283
+ " <td>3.903090</td>\n",
1284
+ " </tr>\n",
1285
+ " <tr>\n",
1286
+ " <th>1849404</th>\n",
1287
+ " <td>KFPRVKNWELGSITYDTLCAQSQQDGPCTPRRCLGSLVLPRKLQTR...</td>\n",
1288
+ " <td>CC[Se]C(=N)N</td>\n",
1289
+ " <td>0.039</td>\n",
1290
+ " <td>7.408935</td>\n",
1291
+ " </tr>\n",
1292
+ " </tbody>\n",
1293
+ "</table>\n",
1294
+ "<p>1849405 rows × 4 columns</p>\n",
1295
+ "</div>"
1296
+ ],
1297
+ "text/plain": [
1298
+ " seq \\\n",
1299
+ "0 APQTITELCSEYRNTQIYTINDKILSYTESMAGKREMVIITFKSGE... \n",
1300
+ "1 VETFAFQAEIAQLMSLIINTFYSNKEIFLRELISNSSDALDKIRYE... \n",
1301
+ "2 GMRVYLGADHAGYELKQRIIEHLKQTGHEPIDCGALRYDADDDYPA... \n",
1302
+ "3 SMENFQKVEKIGEGTYGVVYKARNKLTGEVVALKKIRLDTETEGVP... \n",
1303
+ "4 EFSEWFHNILEEAEIIDQRYPVKGMHVWMPHGFMIRKNTLKILRRI... \n",
1304
+ "... ... \n",
1305
+ "1849400 KQISVRGLAGVENVTELKKNFNRHLHFTLVKDRNVATPRDYYFALA... \n",
1306
+ "1849401 IVEGSDAEIGMSPWQVMLFRKSPQELLCGASLISDRWVLTAAHCLL... \n",
1307
+ "1849402 IVEGSDAEIGMSPWQVMLFRKSPQELLCGASLISDRWVLTAAHCLL... \n",
1308
+ "1849403 SKVVVPAQGKKITLQNGKLNVPENPIIPYIEGDGIGVDVTPAMLKV... \n",
1309
+ "1849404 KFPRVKNWELGSITYDTLCAQSQQDGPCTPRRCLGSLVLPRKLQTR... \n",
1310
+ "\n",
1311
+ " smiles affinity_uM \\\n",
1312
+ "0 OC[C@H]1O[C@H](Oc2cccc(c2)N(=O)=O)[C@@H]([C@H]... 500.000 \n",
1313
+ "1 COc1ccc(cc1)c1c(onc1c1cc(C(C)C)c(cc1O)O)NC(=O)... 0.023 \n",
1314
+ "2 O[C@H]1O[C@H](CO[P](=O)(=O)=O)[C@H]([C@H]([C@H... 6300.000 \n",
1315
+ "3 OCC[C@@H]1CCCCN1c1cc(NCC2=CC=CN(C2)O)n2c(n1)c(... 0.210 \n",
1316
+ "4 O[C@@H]1[C@@H](COS(=O)(=O)NC(=O)[C@@H]2CCC[NH2... 0.050 \n",
1317
+ "... ... ... \n",
1318
+ "1849400 O[C@@H]1[C@H](O)[C@H](O[C@H]1n1cnc2c1ncnc2N)CO... 250.000 \n",
1319
+ "1849401 O=C[C@@H](NC(=O)[C@H](Cc1ccc(cc1)OS(O)(O)O)NC(... 8.000 \n",
1320
+ "1849402 CC(C[C@@H](C(=O)N1C=CC[C@H]1C(=O)N)NC(=O)[C@@H... 8.000 \n",
1321
+ "1849403 O[C@@H]1[C@@H](COP(=O)(O)O)O[C@H]([C@@H]1OP(=O... 125.000 \n",
1322
+ "1849404 CC[Se]C(=N)N 0.039 \n",
1323
+ "\n",
1324
+ " neg_log10_affinity_M \n",
1325
+ "0 3.301030 \n",
1326
+ "1 7.638272 \n",
1327
+ "2 2.200659 \n",
1328
+ "3 6.677781 \n",
1329
+ "4 7.301030 \n",
1330
+ "... ... \n",
1331
+ "1849400 3.602060 \n",
1332
+ "1849401 5.096910 \n",
1333
+ "1849402 5.096910 \n",
1334
+ "1849403 3.903090 \n",
1335
+ "1849404 7.408935 \n",
1336
+ "\n",
1337
+ "[1849405 rows x 4 columns]"
1338
+ ]
1339
+ },
1340
+ "execution_count": 70,
1341
+ "metadata": {},
1342
+ "output_type": "execute_result"
1343
+ }
1344
+ ],
1345
+ "source": [
1346
+ "df_nr"
1347
+ ]
1348
+ },
1349
+ {
1350
+ "cell_type": "code",
1351
+ "execution_count": 72,
1352
+ "id": "7f4027a2-0a5f-47bf-8a34-0c6a73b9b112",
1353
+ "metadata": {},
1354
+ "outputs": [],
1355
+ "source": [
1356
+ "df = df_nr[np.isfinite(df_nr['neg_log10_affinity_M'])]"
1357
+ ]
1358
+ },
1359
+ {
1360
+ "cell_type": "code",
1361
+ "execution_count": 86,
1362
+ "id": "c558f3f6-9fe7-4361-8272-23a54368fdda",
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
+ {
1380
+ "cell_type": "code",
1381
+ "execution_count": 5,
1382
+ "id": "d91c0d91-474c-4ab2-9a5e-3b7861f7a832",
1383
+ "metadata": {},
1384
+ "outputs": [
1385
+ {
1386
+ "data": {
1387
+ "image/png": "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\n",
1388
+ "text/plain": [
1389
+ "<Figure size 432x288 with 1 Axes>"
1390
+ ]
1391
+ },
1392
+ "metadata": {
1393
+ "needs_background": "light"
1394
+ },
1395
+ "output_type": "display_data"
1396
+ }
1397
+ ],
1398
+ "source": [
1399
+ "ax = df['neg_log10_affinity_M'].hist(bins=100,density=True)\n",
1400
+ "ax.set_xlabel('-$\\log_{10}$ affinity[M]',fontsize=16)\n",
1401
+ "ax.set_ylabel('probability',fontsize=16)\n",
1402
+ "ax.figure.savefig('affinity.pdf')"
1403
+ ]
1404
+ },
1405
+ {
1406
+ "cell_type": "code",
1407
+ "execution_count": 6,
1408
+ "id": "11571486-901c-474b-a8ec-215ec5c9e661",
1409
+ "metadata": {},
1410
+ "outputs": [
1411
+ {
1412
+ "data": {
1413
+ "text/plain": [
1414
+ "1848949"
1415
+ ]
1416
+ },
1417
+ "execution_count": 6,
1418
+ "metadata": {},
1419
+ "output_type": "execute_result"
1420
+ }
1421
+ ],
1422
+ "source": [
1423
+ "len(df)"
1424
+ ]
1425
+ },
1426
+ {
1427
+ "cell_type": "code",
1428
+ "execution_count": 7,
1429
+ "id": "9ca8df46-15d3-40f9-b304-dd6e5597be5e",
1430
+ "metadata": {},
1431
+ "outputs": [
1432
+ {
1433
+ "data": {
1434
+ "text/plain": [
1435
+ "5.142857142857143"
1436
+ ]
1437
+ },
1438
+ "execution_count": 7,
1439
+ "metadata": {},
1440
+ "output_type": "execute_result"
1441
+ }
1442
+ ],
1443
+ "source": [
1444
+ "1.8/0.35"
1445
+ ]
1446
+ },
1447
+ {
1448
+ "cell_type": "code",
1449
+ "execution_count": null,
1450
+ "id": "88cf855d-704f-4ed4-827e-9f4e3288b3a0",
1451
+ "metadata": {},
1452
+ "outputs": [],
1453
+ "source": []
1454
+ }
1455
+ ],
1456
+ "metadata": {
1457
+ "kernelspec": {
1458
+ "display_name": "Python 3",
1459
+ "language": "python",
1460
+ "name": "python3"
1461
+ },
1462
+ "language_info": {
1463
+ "codemirror_mode": {
1464
+ "name": "ipython",
1465
+ "version": 3
1466
+ },
1467
+ "file_extension": ".py",
1468
+ "mimetype": "text/x-python",
1469
+ "name": "python",
1470
+ "nbconvert_exporter": "python",
1471
+ "pygments_lexer": "ipython3",
1472
+ "version": "3.9.4"
1473
+ }
1474
+ },
1475
+ "nbformat": 4,
1476
+ "nbformat_minor": 5
1477
+ }
moad.ipynb ADDED
@@ -0,0 +1,513 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": 3,
6
+ "id": "c47a32d8-c857-41de-a70a-cec48046df12",
7
+ "metadata": {},
8
+ "outputs": [],
9
+ "source": [
10
+ "import pandas as pd"
11
+ ]
12
+ },
13
+ {
14
+ "cell_type": "code",
15
+ "execution_count": 92,
16
+ "id": "e0c6bd53-3417-44bd-b1b4-81802b37fbfc",
17
+ "metadata": {},
18
+ "outputs": [],
19
+ "source": [
20
+ "df = pd.read_csv('binding_moad/every.csv',header=None,skiprows=2)\n",
21
+ "df = df.rename(columns={2:'pdb',3: 'ligand_name', 4: 'ligand_valid', 7: 'affinity_val', 8: 'affinity_unit', 9:'smiles'})\n",
22
+ "#df = df[df['ligand_valid']!='invalid'].copy()"
23
+ ]
24
+ },
25
+ {
26
+ "cell_type": "code",
27
+ "execution_count": 93,
28
+ "id": "e40b1ddc-9a98-4a3b-b8a6-45e3940a3ea2",
29
+ "metadata": {},
30
+ "outputs": [],
31
+ "source": [
32
+ "df['is_sep'] = df[1] == 'Family. Representative Entry is '"
33
+ ]
34
+ },
35
+ {
36
+ "cell_type": "code",
37
+ "execution_count": 94,
38
+ "id": "4f00a0d1-78db-4f32-9d12-5e035b70ef98",
39
+ "metadata": {},
40
+ "outputs": [],
41
+ "source": [
42
+ "df['cum_sum'] = df['is_sep'].cumsum()"
43
+ ]
44
+ },
45
+ {
46
+ "cell_type": "code",
47
+ "execution_count": 95,
48
+ "id": "52c0c66c-1eb0-415b-b019-bc77419ccbd7",
49
+ "metadata": {},
50
+ "outputs": [],
51
+ "source": [
52
+ "from pint import UnitRegistry\n",
53
+ "ureg = UnitRegistry()\n",
54
+ "\n",
55
+ "def to_uM(affinity_unit):\n",
56
+ " try:\n",
57
+ " val = ureg(str(affinity_unit[0])+str(affinity_unit[1]))\n",
58
+ " return val.m_as(ureg.uM)\n",
59
+ " except Exception:\n",
60
+ " pass\n",
61
+ " \n",
62
+ " try:\n",
63
+ " val = ureg(str(affinity_unit[0])+str(affinity_unit[1]))\n",
64
+ " return 1/val.m_as(1/ureg.uM)\n",
65
+ " except Exception:\n",
66
+ " pass"
67
+ ]
68
+ },
69
+ {
70
+ "cell_type": "code",
71
+ "execution_count": 96,
72
+ "id": "e5b4dd41-1389-408d-bee6-6dbeefc1d5c7",
73
+ "metadata": {},
74
+ "outputs": [],
75
+ "source": [
76
+ "groupby = df.groupby('cum_sum')"
77
+ ]
78
+ },
79
+ {
80
+ "cell_type": "code",
81
+ "execution_count": 121,
82
+ "id": "61b8276c-54fe-4989-af5f-723994e1df7e",
83
+ "metadata": {},
84
+ "outputs": [],
85
+ "source": [
86
+ "def group(df):\n",
87
+ " pdb = df[df['is_sep']]['pdb'].values\n",
88
+ " if len(pdb) > 0:\n",
89
+ " pdb = pdb[0]\n",
90
+ " df['pdb_ref'] = pdb\n",
91
+ " return df[df['ligand_valid']=='valid']\n",
92
+ "df_expand = groupby.apply(group).reset_index(drop=True)"
93
+ ]
94
+ },
95
+ {
96
+ "cell_type": "code",
97
+ "execution_count": 124,
98
+ "id": "8bb2dfac-5f11-455c-9dee-3607b47b4232",
99
+ "metadata": {},
100
+ "outputs": [],
101
+ "source": [
102
+ "df_expand['affinity_uM'] = df_expand[['affinity_val','affinity_unit']].apply(to_uM,axis=1)"
103
+ ]
104
+ },
105
+ {
106
+ "cell_type": "code",
107
+ "execution_count": 125,
108
+ "id": "0dc39f62-5b18-4a86-9a44-17d1925da2ad",
109
+ "metadata": {},
110
+ "outputs": [],
111
+ "source": [
112
+ "df_complex = pd.read_parquet('data/moad_complex.parquet')\n",
113
+ "df_complex['name'] = df_complex['name'].str.upper()"
114
+ ]
115
+ },
116
+ {
117
+ "cell_type": "code",
118
+ "execution_count": 128,
119
+ "id": "6d158a41-64c6-4fa2-92d5-562aa11e8924",
120
+ "metadata": {},
121
+ "outputs": [],
122
+ "source": [
123
+ "df_all = df_expand.merge(df_complex,left_on='pdb_ref',right_on='name')"
124
+ ]
125
+ },
126
+ {
127
+ "cell_type": "code",
128
+ "execution_count": 129,
129
+ "id": "901fe6c6-dc8c-4ce4-82c6-1fb0b718287a",
130
+ "metadata": {},
131
+ "outputs": [],
132
+ "source": [
133
+ "df_all = df_all[~df_all['affinity_val'].isnull()]"
134
+ ]
135
+ },
136
+ {
137
+ "cell_type": "code",
138
+ "execution_count": 130,
139
+ "id": "383f9a1c-ffc6-43da-ac5a-5bcb815be28b",
140
+ "metadata": {},
141
+ "outputs": [
142
+ {
143
+ "data": {
144
+ "text/html": [
145
+ "<div>\n",
146
+ "<style scoped>\n",
147
+ " .dataframe tbody tr th:only-of-type {\n",
148
+ " vertical-align: middle;\n",
149
+ " }\n",
150
+ "\n",
151
+ " .dataframe tbody tr th {\n",
152
+ " vertical-align: top;\n",
153
+ " }\n",
154
+ "\n",
155
+ " .dataframe thead th {\n",
156
+ " text-align: right;\n",
157
+ " }\n",
158
+ "</style>\n",
159
+ "<table border=\"1\" class=\"dataframe\">\n",
160
+ " <thead>\n",
161
+ " <tr style=\"text-align: right;\">\n",
162
+ " <th></th>\n",
163
+ " <th>0</th>\n",
164
+ " <th>1</th>\n",
165
+ " <th>pdb</th>\n",
166
+ " <th>ligand_name</th>\n",
167
+ " <th>ligand_valid</th>\n",
168
+ " <th>5</th>\n",
169
+ " <th>6</th>\n",
170
+ " <th>affinity_val</th>\n",
171
+ " <th>affinity_unit</th>\n",
172
+ " <th>smiles</th>\n",
173
+ " <th>10</th>\n",
174
+ " <th>is_sep</th>\n",
175
+ " <th>cum_sum</th>\n",
176
+ " <th>pdb_ref</th>\n",
177
+ " <th>affinity_uM</th>\n",
178
+ " <th>name</th>\n",
179
+ " <th>seq</th>\n",
180
+ " </tr>\n",
181
+ " </thead>\n",
182
+ " <tbody>\n",
183
+ " <tr>\n",
184
+ " <th>0</th>\n",
185
+ " <td>NaN</td>\n",
186
+ " <td>NaN</td>\n",
187
+ " <td>NaN</td>\n",
188
+ " <td>2PA:C:613</td>\n",
189
+ " <td>valid</td>\n",
190
+ " <td>Ki</td>\n",
191
+ " <td>=</td>\n",
192
+ " <td>0.62</td>\n",
193
+ " <td>nM</td>\n",
194
+ " <td>NP(=O)(N)O</td>\n",
195
+ " <td>NaN</td>\n",
196
+ " <td>False</td>\n",
197
+ " <td>1</td>\n",
198
+ " <td>6H8J</td>\n",
199
+ " <td>0.000620</td>\n",
200
+ " <td>6H8J</td>\n",
201
+ " <td>NYIVPGEYRVAEGEIEINAGREKTTIRVSNTGDRPIQVGSHIHFVE...</td>\n",
202
+ " </tr>\n",
203
+ " <tr>\n",
204
+ " <th>2</th>\n",
205
+ " <td>NaN</td>\n",
206
+ " <td>NaN</td>\n",
207
+ " <td>NaN</td>\n",
208
+ " <td>HAE:C:800</td>\n",
209
+ " <td>valid</td>\n",
210
+ " <td>Ki</td>\n",
211
+ " <td>=</td>\n",
212
+ " <td>2.60</td>\n",
213
+ " <td>uM</td>\n",
214
+ " <td>CC(=O)NO</td>\n",
215
+ " <td>NaN</td>\n",
216
+ " <td>False</td>\n",
217
+ " <td>1</td>\n",
218
+ " <td>6H8J</td>\n",
219
+ " <td>2.600000</td>\n",
220
+ " <td>6H8J</td>\n",
221
+ " <td>NYIVPGEYRVAEGEIEINAGREKTTIRVSNTGDRPIQVGSHIHFVE...</td>\n",
222
+ " </tr>\n",
223
+ " <tr>\n",
224
+ " <th>7</th>\n",
225
+ " <td>NaN</td>\n",
226
+ " <td>NaN</td>\n",
227
+ " <td>NaN</td>\n",
228
+ " <td>43W:A:902</td>\n",
229
+ " <td>valid</td>\n",
230
+ " <td>ic50</td>\n",
231
+ " <td>=</td>\n",
232
+ " <td>580.00</td>\n",
233
+ " <td>nM</td>\n",
234
+ " <td>C#CCCOP(=O)(O)OP(=O)(O)O</td>\n",
235
+ " <td>NaN</td>\n",
236
+ " <td>False</td>\n",
237
+ " <td>2</td>\n",
238
+ " <td>4S3F</td>\n",
239
+ " <td>0.580000</td>\n",
240
+ " <td>4S3F</td>\n",
241
+ " <td>MEGMRRPTPTVYVGRVPIGGAHPIAVQSMTNTPTRDVEATTAQVLE...</td>\n",
242
+ " </tr>\n",
243
+ " <tr>\n",
244
+ " <th>16</th>\n",
245
+ " <td>NaN</td>\n",
246
+ " <td>NaN</td>\n",
247
+ " <td>NaN</td>\n",
248
+ " <td>0CG:A:902</td>\n",
249
+ " <td>valid</td>\n",
250
+ " <td>ic50</td>\n",
251
+ " <td>=</td>\n",
252
+ " <td>770.00</td>\n",
253
+ " <td>nM</td>\n",
254
+ " <td>C#CCOP(=O)(O)OP(=O)(O)O</td>\n",
255
+ " <td>NaN</td>\n",
256
+ " <td>False</td>\n",
257
+ " <td>2</td>\n",
258
+ " <td>4S3F</td>\n",
259
+ " <td>0.770000</td>\n",
260
+ " <td>4S3F</td>\n",
261
+ " <td>MEGMRRPTPTVYVGRVPIGGAHPIAVQSMTNTPTRDVEATTAQVLE...</td>\n",
262
+ " </tr>\n",
263
+ " <tr>\n",
264
+ " <th>17</th>\n",
265
+ " <td>NaN</td>\n",
266
+ " <td>NaN</td>\n",
267
+ " <td>NaN</td>\n",
268
+ " <td>ADN:A:901</td>\n",
269
+ " <td>valid</td>\n",
270
+ " <td>Kd</td>\n",
271
+ " <td>=</td>\n",
272
+ " <td>15.00</td>\n",
273
+ " <td>uM</td>\n",
274
+ " <td>c1nc(c2c(n1)n(cn2)[C@H]3[C@@H]([C@@H]([C@H](O3...</td>\n",
275
+ " <td>NaN</td>\n",
276
+ " <td>False</td>\n",
277
+ " <td>5</td>\n",
278
+ " <td>2GL0</td>\n",
279
+ " <td>15.000000</td>\n",
280
+ " <td>2GL0</td>\n",
281
+ " <td>MTDMSIKFELIDVPIPQGTNVIIGQAHFIKTVEDLYEALVTSVPGV...</td>\n",
282
+ " </tr>\n",
283
+ " <tr>\n",
284
+ " <th>...</th>\n",
285
+ " <td>...</td>\n",
286
+ " <td>...</td>\n",
287
+ " <td>...</td>\n",
288
+ " <td>...</td>\n",
289
+ " <td>...</td>\n",
290
+ " <td>...</td>\n",
291
+ " <td>...</td>\n",
292
+ " <td>...</td>\n",
293
+ " <td>...</td>\n",
294
+ " <td>...</td>\n",
295
+ " <td>...</td>\n",
296
+ " <td>...</td>\n",
297
+ " <td>...</td>\n",
298
+ " <td>...</td>\n",
299
+ " <td>...</td>\n",
300
+ " <td>...</td>\n",
301
+ " <td>...</td>\n",
302
+ " </tr>\n",
303
+ " <tr>\n",
304
+ " <th>51900</th>\n",
305
+ " <td>NaN</td>\n",
306
+ " <td>NaN</td>\n",
307
+ " <td>NaN</td>\n",
308
+ " <td>MAN NAG:G:1</td>\n",
309
+ " <td>valid</td>\n",
310
+ " <td>Ka</td>\n",
311
+ " <td>=</td>\n",
312
+ " <td>7860.00</td>\n",
313
+ " <td>M^-1</td>\n",
314
+ " <td>NaN</td>\n",
315
+ " <td>NaN</td>\n",
316
+ " <td>False</td>\n",
317
+ " <td>10499</td>\n",
318
+ " <td>2WDB</td>\n",
319
+ " <td>127.226463</td>\n",
320
+ " <td>2WDB</td>\n",
321
+ " <td>MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG...</td>\n",
322
+ " </tr>\n",
323
+ " <tr>\n",
324
+ " <th>51901</th>\n",
325
+ " <td>NaN</td>\n",
326
+ " <td>NaN</td>\n",
327
+ " <td>NaN</td>\n",
328
+ " <td>MAN NAG:F:1</td>\n",
329
+ " <td>valid</td>\n",
330
+ " <td>Ka</td>\n",
331
+ " <td>=</td>\n",
332
+ " <td>7860.00</td>\n",
333
+ " <td>M^-1</td>\n",
334
+ " <td>NaN</td>\n",
335
+ " <td>NaN</td>\n",
336
+ " <td>False</td>\n",
337
+ " <td>10499</td>\n",
338
+ " <td>2WDB</td>\n",
339
+ " <td>127.226463</td>\n",
340
+ " <td>2WDB</td>\n",
341
+ " <td>MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG...</td>\n",
342
+ " </tr>\n",
343
+ " <tr>\n",
344
+ " <th>51902</th>\n",
345
+ " <td>NaN</td>\n",
346
+ " <td>NaN</td>\n",
347
+ " <td>NaN</td>\n",
348
+ " <td>NGA NAG:F:1</td>\n",
349
+ " <td>valid</td>\n",
350
+ " <td>Ka</td>\n",
351
+ " <td>=</td>\n",
352
+ " <td>5910.00</td>\n",
353
+ " <td>M^-1</td>\n",
354
+ " <td>NaN</td>\n",
355
+ " <td>NaN</td>\n",
356
+ " <td>False</td>\n",
357
+ " <td>10499</td>\n",
358
+ " <td>2WDB</td>\n",
359
+ " <td>169.204738</td>\n",
360
+ " <td>2WDB</td>\n",
361
+ " <td>MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG...</td>\n",
362
+ " </tr>\n",
363
+ " <tr>\n",
364
+ " <th>51903</th>\n",
365
+ " <td>NaN</td>\n",
366
+ " <td>NaN</td>\n",
367
+ " <td>NaN</td>\n",
368
+ " <td>NGA NAG:E:1</td>\n",
369
+ " <td>valid</td>\n",
370
+ " <td>Ka</td>\n",
371
+ " <td>=</td>\n",
372
+ " <td>5910.00</td>\n",
373
+ " <td>M^-1</td>\n",
374
+ " <td>NaN</td>\n",
375
+ " <td>NaN</td>\n",
376
+ " <td>False</td>\n",
377
+ " <td>10499</td>\n",
378
+ " <td>2WDB</td>\n",
379
+ " <td>169.204738</td>\n",
380
+ " <td>2WDB</td>\n",
381
+ " <td>MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG...</td>\n",
382
+ " </tr>\n",
383
+ " <tr>\n",
384
+ " <th>51904</th>\n",
385
+ " <td>NaN</td>\n",
386
+ " <td>NaN</td>\n",
387
+ " <td>NaN</td>\n",
388
+ " <td>NGA NAG:H:1</td>\n",
389
+ " <td>valid</td>\n",
390
+ " <td>Ka</td>\n",
391
+ " <td>=</td>\n",
392
+ " <td>5910.00</td>\n",
393
+ " <td>M^-1</td>\n",
394
+ " <td>NaN</td>\n",
395
+ " <td>NaN</td>\n",
396
+ " <td>False</td>\n",
397
+ " <td>10499</td>\n",
398
+ " <td>2WDB</td>\n",
399
+ " <td>169.204738</td>\n",
400
+ " <td>2WDB</td>\n",
401
+ " <td>MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG...</td>\n",
402
+ " </tr>\n",
403
+ " </tbody>\n",
404
+ "</table>\n",
405
+ "<p>25425 rows × 17 columns</p>\n",
406
+ "</div>"
407
+ ],
408
+ "text/plain": [
409
+ " 0 1 pdb ligand_name ligand_valid 5 6 affinity_val \\\n",
410
+ "0 NaN NaN NaN 2PA:C:613 valid Ki = 0.62 \n",
411
+ "2 NaN NaN NaN HAE:C:800 valid Ki = 2.60 \n",
412
+ "7 NaN NaN NaN 43W:A:902 valid ic50 = 580.00 \n",
413
+ "16 NaN NaN NaN 0CG:A:902 valid ic50 = 770.00 \n",
414
+ "17 NaN NaN NaN ADN:A:901 valid Kd = 15.00 \n",
415
+ "... ... ... ... ... ... ... .. ... \n",
416
+ "51900 NaN NaN NaN MAN NAG:G:1 valid Ka = 7860.00 \n",
417
+ "51901 NaN NaN NaN MAN NAG:F:1 valid Ka = 7860.00 \n",
418
+ "51902 NaN NaN NaN NGA NAG:F:1 valid Ka = 5910.00 \n",
419
+ "51903 NaN NaN NaN NGA NAG:E:1 valid Ka = 5910.00 \n",
420
+ "51904 NaN NaN NaN NGA NAG:H:1 valid Ka = 5910.00 \n",
421
+ "\n",
422
+ " affinity_unit smiles 10 \\\n",
423
+ "0 nM NP(=O)(N)O NaN \n",
424
+ "2 uM CC(=O)NO NaN \n",
425
+ "7 nM C#CCCOP(=O)(O)OP(=O)(O)O NaN \n",
426
+ "16 nM C#CCOP(=O)(O)OP(=O)(O)O NaN \n",
427
+ "17 uM c1nc(c2c(n1)n(cn2)[C@H]3[C@@H]([C@@H]([C@H](O3... NaN \n",
428
+ "... ... ... .. \n",
429
+ "51900 M^-1 NaN NaN \n",
430
+ "51901 M^-1 NaN NaN \n",
431
+ "51902 M^-1 NaN NaN \n",
432
+ "51903 M^-1 NaN NaN \n",
433
+ "51904 M^-1 NaN NaN \n",
434
+ "\n",
435
+ " is_sep cum_sum pdb_ref affinity_uM name \\\n",
436
+ "0 False 1 6H8J 0.000620 6H8J \n",
437
+ "2 False 1 6H8J 2.600000 6H8J \n",
438
+ "7 False 2 4S3F 0.580000 4S3F \n",
439
+ "16 False 2 4S3F 0.770000 4S3F \n",
440
+ "17 False 5 2GL0 15.000000 2GL0 \n",
441
+ "... ... ... ... ... ... \n",
442
+ "51900 False 10499 2WDB 127.226463 2WDB \n",
443
+ "51901 False 10499 2WDB 127.226463 2WDB \n",
444
+ "51902 False 10499 2WDB 169.204738 2WDB \n",
445
+ "51903 False 10499 2WDB 169.204738 2WDB \n",
446
+ "51904 False 10499 2WDB 169.204738 2WDB \n",
447
+ "\n",
448
+ " seq \n",
449
+ "0 NYIVPGEYRVAEGEIEINAGREKTTIRVSNTGDRPIQVGSHIHFVE... \n",
450
+ "2 NYIVPGEYRVAEGEIEINAGREKTTIRVSNTGDRPIQVGSHIHFVE... \n",
451
+ "7 MEGMRRPTPTVYVGRVPIGGAHPIAVQSMTNTPTRDVEATTAQVLE... \n",
452
+ "16 MEGMRRPTPTVYVGRVPIGGAHPIAVQSMTNTPTRDVEATTAQVLE... \n",
453
+ "17 MTDMSIKFELIDVPIPQGTNVIIGQAHFIKTVEDLYEALVTSVPGV... \n",
454
+ "... ... \n",
455
+ "51900 MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG... \n",
456
+ "51901 MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG... \n",
457
+ "51902 MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG... \n",
458
+ "51903 MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG... \n",
459
+ "51904 MGSSHHHHHHSSGLVPRGSHMASNPSLIRSESWQVYEGNEANLLDG... \n",
460
+ "\n",
461
+ "[25425 rows x 17 columns]"
462
+ ]
463
+ },
464
+ "execution_count": 130,
465
+ "metadata": {},
466
+ "output_type": "execute_result"
467
+ }
468
+ ],
469
+ "source": [
470
+ "df_all"
471
+ ]
472
+ },
473
+ {
474
+ "cell_type": "code",
475
+ "execution_count": 133,
476
+ "id": "bebc962b-10f7-478c-8e23-e2d3722e875c",
477
+ "metadata": {},
478
+ "outputs": [],
479
+ "source": [
480
+ "df_all[['pdb','ligand_name','smiles','name','affinity_uM','seq']].to_parquet('data/moad.parquet')"
481
+ ]
482
+ },
483
+ {
484
+ "cell_type": "code",
485
+ "execution_count": null,
486
+ "id": "6ceb8706-273c-4a83-8cda-c7e33fc87e38",
487
+ "metadata": {},
488
+ "outputs": [],
489
+ "source": []
490
+ }
491
+ ],
492
+ "metadata": {
493
+ "kernelspec": {
494
+ "display_name": "Python 3",
495
+ "language": "python",
496
+ "name": "python3"
497
+ },
498
+ "language_info": {
499
+ "codemirror_mode": {
500
+ "name": "ipython",
501
+ "version": 3
502
+ },
503
+ "file_extension": ".py",
504
+ "mimetype": "text/x-python",
505
+ "name": "python",
506
+ "nbconvert_exporter": "python",
507
+ "pygments_lexer": "ipython3",
508
+ "version": "3.9.4"
509
+ }
510
+ },
511
+ "nbformat": 4,
512
+ "nbformat_minor": 5
513
+ }
moad.py ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from mpi4py import MPI
2
+ from mpi4py.futures import MPICommExecutor
3
+
4
+ from openbabel import pybel
5
+ from Bio import SeqIO
6
+
7
+ import os
8
+ def parse_complex(fn):
9
+ try:
10
+ name = os.path.basename(fn).split('.')[0]
11
+ print(name)
12
+ seq = str(next(SeqIO.parse(fn, "pdb-seqres")).seq)
13
+ return name, seq
14
+ except:
15
+ return None
16
+
17
+
18
+ if __name__ == '__main__':
19
+ import glob
20
+
21
+ filenames = glob.glob('binding_moad/BindingMOAD_2020/*.bio1')
22
+ comm = MPI.COMM_WORLD
23
+ with MPICommExecutor(comm, root=0) as executor:
24
+ if executor is not None:
25
+ result = executor.map(parse_complex, filenames)
26
+ result = list(result)
27
+ names = [r[0] for r in result if r is not None]
28
+ seqs = [r[1] for r in result if r is not None]
29
+
30
+ import pandas as pd
31
+ df = pd.DataFrame({'name': names, 'seq': seqs})
32
+ df.to_parquet('data/moad_complex.parquet')
pdbbind.ipynb ADDED
@@ -0,0 +1,296 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "markdown",
5
+ "id": "834aeced-c3c5-42a0-bad1-41e009dd86ee",
6
+ "metadata": {},
7
+ "source": [
8
+ "### Preprocessing"
9
+ ]
10
+ },
11
+ {
12
+ "cell_type": "code",
13
+ "execution_count": 16,
14
+ "id": "86476f6e-802a-463b-a1b0-2ae228bb92af",
15
+ "metadata": {},
16
+ "outputs": [],
17
+ "source": [
18
+ "import pandas as pd"
19
+ ]
20
+ },
21
+ {
22
+ "cell_type": "code",
23
+ "execution_count": null,
24
+ "id": "0cde27df-2f77-4e62-8c65-7b7a4e76b404",
25
+ "metadata": {},
26
+ "outputs": [],
27
+ "source": [
28
+ "complex = pd.read_parquet('')"
29
+ ]
30
+ },
31
+ {
32
+ "cell_type": "code",
33
+ "execution_count": 49,
34
+ "id": "9b2be11c-f4bb-4107-af49-abd78052afcf",
35
+ "metadata": {},
36
+ "outputs": [],
37
+ "source": [
38
+ "df = pd.read_table('pdbbind/data/plain-text-index/index/INDEX_general_PL_data.2019',skiprows=4,sep=r'\\s+',usecols=[0,4]).drop(0)\n",
39
+ "df = df.rename(columns={'#': 'name','release': 'affinity'})"
40
+ ]
41
+ },
42
+ {
43
+ "cell_type": "code",
44
+ "execution_count": 50,
45
+ "id": "16e0fe44-96aa-4d3a-ae42-3609e895418b",
46
+ "metadata": {},
47
+ "outputs": [],
48
+ "source": [
49
+ "from numericalunits import mL, nm"
50
+ ]
51
+ },
52
+ {
53
+ "cell_type": "code",
54
+ "execution_count": 136,
55
+ "id": "3acbca3c-9c0b-43a1-a45e-331bf153bcfa",
56
+ "metadata": {},
57
+ "outputs": [],
58
+ "source": [
59
+ "from pint import UnitRegistry\n",
60
+ "ureg = UnitRegistry()\n",
61
+ "\n",
62
+ "def to_uM(affinity):\n",
63
+ " val = ureg(affinity)\n",
64
+ " try:\n",
65
+ " return val.m_as(ureg.uM)\n",
66
+ " except Exception:\n",
67
+ " pass\n",
68
+ " \n",
69
+ " try:\n",
70
+ " return 1/val.m_as(1/ureg.uM)\n",
71
+ " except Exception:\n",
72
+ " pass"
73
+ ]
74
+ },
75
+ {
76
+ "cell_type": "code",
77
+ "execution_count": 137,
78
+ "id": "58e5748b-2cea-43ff-ab51-85a5021bd50b",
79
+ "metadata": {},
80
+ "outputs": [],
81
+ "source": [
82
+ "df['affinity_uM'] = df['affinity'].str.split('[=\\~><]').str[1].apply(to_uM)"
83
+ ]
84
+ },
85
+ {
86
+ "cell_type": "code",
87
+ "execution_count": 138,
88
+ "id": "d92f0004-68c1-4487-94b9-56b4fd598de4",
89
+ "metadata": {},
90
+ "outputs": [
91
+ {
92
+ "data": {
93
+ "text/html": [
94
+ "<div>\n",
95
+ "<style scoped>\n",
96
+ " .dataframe tbody tr th:only-of-type {\n",
97
+ " vertical-align: middle;\n",
98
+ " }\n",
99
+ "\n",
100
+ " .dataframe tbody tr th {\n",
101
+ " vertical-align: top;\n",
102
+ " }\n",
103
+ "\n",
104
+ " .dataframe thead th {\n",
105
+ " text-align: right;\n",
106
+ " }\n",
107
+ "</style>\n",
108
+ "<table border=\"1\" class=\"dataframe\">\n",
109
+ " <thead>\n",
110
+ " <tr style=\"text-align: right;\">\n",
111
+ " <th></th>\n",
112
+ " <th>name</th>\n",
113
+ " <th>affinity</th>\n",
114
+ " <th>affinity_uM</th>\n",
115
+ " </tr>\n",
116
+ " </thead>\n",
117
+ " <tbody>\n",
118
+ " <tr>\n",
119
+ " <th>1</th>\n",
120
+ " <td>3zzf</td>\n",
121
+ " <td>Ki=400mM</td>\n",
122
+ " <td>4.000000e+05</td>\n",
123
+ " </tr>\n",
124
+ " <tr>\n",
125
+ " <th>2</th>\n",
126
+ " <td>3gww</td>\n",
127
+ " <td>IC50=355mM</td>\n",
128
+ " <td>3.550000e+05</td>\n",
129
+ " </tr>\n",
130
+ " <tr>\n",
131
+ " <th>3</th>\n",
132
+ " <td>1w8l</td>\n",
133
+ " <td>Ki=320mM</td>\n",
134
+ " <td>3.200000e+05</td>\n",
135
+ " </tr>\n",
136
+ " <tr>\n",
137
+ " <th>4</th>\n",
138
+ " <td>3fqa</td>\n",
139
+ " <td>IC50=320mM</td>\n",
140
+ " <td>3.200000e+05</td>\n",
141
+ " </tr>\n",
142
+ " <tr>\n",
143
+ " <th>5</th>\n",
144
+ " <td>1zsb</td>\n",
145
+ " <td>Kd=250mM</td>\n",
146
+ " <td>2.500000e+05</td>\n",
147
+ " </tr>\n",
148
+ " <tr>\n",
149
+ " <th>...</th>\n",
150
+ " <td>...</td>\n",
151
+ " <td>...</td>\n",
152
+ " <td>...</td>\n",
153
+ " </tr>\n",
154
+ " <tr>\n",
155
+ " <th>17675</th>\n",
156
+ " <td>7cpa</td>\n",
157
+ " <td>Ki=11fM</td>\n",
158
+ " <td>1.100000e-08</td>\n",
159
+ " </tr>\n",
160
+ " <tr>\n",
161
+ " <th>17676</th>\n",
162
+ " <td>2xuf</td>\n",
163
+ " <td>Kd=4.1fM</td>\n",
164
+ " <td>4.100000e-09</td>\n",
165
+ " </tr>\n",
166
+ " <tr>\n",
167
+ " <th>17677</th>\n",
168
+ " <td>1avd</td>\n",
169
+ " <td>Kd=1fM</td>\n",
170
+ " <td>1.000000e-09</td>\n",
171
+ " </tr>\n",
172
+ " <tr>\n",
173
+ " <th>17678</th>\n",
174
+ " <td>2xui</td>\n",
175
+ " <td>Kd=1.0fM</td>\n",
176
+ " <td>1.000000e-09</td>\n",
177
+ " </tr>\n",
178
+ " <tr>\n",
179
+ " <th>17679</th>\n",
180
+ " <td>2avi</td>\n",
181
+ " <td>Kd=0.6fM</td>\n",
182
+ " <td>6.000000e-10</td>\n",
183
+ " </tr>\n",
184
+ " </tbody>\n",
185
+ "</table>\n",
186
+ "<p>17679 rows × 3 columns</p>\n",
187
+ "</div>"
188
+ ],
189
+ "text/plain": [
190
+ " name affinity affinity_uM\n",
191
+ "1 3zzf Ki=400mM 4.000000e+05\n",
192
+ "2 3gww IC50=355mM 3.550000e+05\n",
193
+ "3 1w8l Ki=320mM 3.200000e+05\n",
194
+ "4 3fqa IC50=320mM 3.200000e+05\n",
195
+ "5 1zsb Kd=250mM 2.500000e+05\n",
196
+ "... ... ... ...\n",
197
+ "17675 7cpa Ki=11fM 1.100000e-08\n",
198
+ "17676 2xuf Kd=4.1fM 4.100000e-09\n",
199
+ "17677 1avd Kd=1fM 1.000000e-09\n",
200
+ "17678 2xui Kd=1.0fM 1.000000e-09\n",
201
+ "17679 2avi Kd=0.6fM 6.000000e-10\n",
202
+ "\n",
203
+ "[17679 rows x 3 columns]"
204
+ ]
205
+ },
206
+ "execution_count": 138,
207
+ "metadata": {},
208
+ "output_type": "execute_result"
209
+ }
210
+ ],
211
+ "source": [
212
+ "df"
213
+ ]
214
+ },
215
+ {
216
+ "cell_type": "code",
217
+ "execution_count": 139,
218
+ "id": "d6dda488-f709-4fe7-b372-080042cf7c66",
219
+ "metadata": {},
220
+ "outputs": [],
221
+ "source": [
222
+ "df_complex = pd.read_parquet('data/pdbbind_complex.parquet')"
223
+ ]
224
+ },
225
+ {
226
+ "cell_type": "code",
227
+ "execution_count": 140,
228
+ "id": "df7929e3-c7fd-4e1b-a165-92f8d53b9011",
229
+ "metadata": {},
230
+ "outputs": [],
231
+ "source": [
232
+ "df_all = df_complex.merge(df,on='name').drop('affinity',axis=1)"
233
+ ]
234
+ },
235
+ {
236
+ "cell_type": "code",
237
+ "execution_count": 141,
238
+ "id": "4d105c42-0d11-49db-9012-52fafc9cd299",
239
+ "metadata": {},
240
+ "outputs": [],
241
+ "source": [
242
+ "df_all.to_parquet('data/pdbbind.parquet')"
243
+ ]
244
+ },
245
+ {
246
+ "cell_type": "code",
247
+ "execution_count": 142,
248
+ "id": "2955b056-26dd-45fa-8d74-f17661253a9a",
249
+ "metadata": {},
250
+ "outputs": [
251
+ {
252
+ "data": {
253
+ "text/plain": [
254
+ "17652"
255
+ ]
256
+ },
257
+ "execution_count": 142,
258
+ "metadata": {},
259
+ "output_type": "execute_result"
260
+ }
261
+ ],
262
+ "source": [
263
+ "len(df_all)"
264
+ ]
265
+ },
266
+ {
267
+ "cell_type": "code",
268
+ "execution_count": null,
269
+ "id": "ed3fe035-6035-4d39-b072-d12dc0a95857",
270
+ "metadata": {},
271
+ "outputs": [],
272
+ "source": []
273
+ }
274
+ ],
275
+ "metadata": {
276
+ "kernelspec": {
277
+ "display_name": "Python 3",
278
+ "language": "python",
279
+ "name": "python3"
280
+ },
281
+ "language_info": {
282
+ "codemirror_mode": {
283
+ "name": "ipython",
284
+ "version": 3
285
+ },
286
+ "file_extension": ".py",
287
+ "mimetype": "text/x-python",
288
+ "name": "python",
289
+ "nbconvert_exporter": "python",
290
+ "pygments_lexer": "ipython3",
291
+ "version": "3.9.4"
292
+ }
293
+ },
294
+ "nbformat": 4,
295
+ "nbformat_minor": 5
296
+ }
pdbbind.py ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from mpi4py import MPI
2
+ from mpi4py.futures import MPICommExecutor
3
+
4
+ from openbabel import pybel
5
+ from Bio import SeqIO
6
+
7
+ import os
8
+ def parse_complex(fn):
9
+ try:
10
+ name = os.path.basename(fn)
11
+ seq = str(next(SeqIO.parse(fn+'/'+name+'_protein.pdb', "pdb-seqres")).seq)
12
+ mol = next(pybel.readfile('sdf',fn+'/'+name+'_ligand.sdf'))
13
+ smi = mol.write('can').split('\t')[0]
14
+ return name, seq, smi
15
+ except:
16
+ return None
17
+
18
+
19
+ if __name__ == '__main__':
20
+ import glob
21
+
22
+ filenames = glob.glob('pdbbind/data/v2019-other-PL/*')
23
+ filenames.extend(glob.glob('pdbbind/data/refined-set/*'))
24
+ comm = MPI.COMM_WORLD
25
+ with MPICommExecutor(comm, root=0) as executor:
26
+ if executor is not None:
27
+ result = executor.map(parse_complex, filenames)
28
+ result = list(result)
29
+ names = [r[0] for r in result if r is not None]
30
+ seqs = [r[1] for r in result if r is not None]
31
+ all_smiles = [r[2] for r in result if r is not None]
32
+
33
+ import pandas as pd
34
+ df = pd.DataFrame({'name': names, 'seq': seqs, 'smiles': all_smiles})
35
+ df.to_parquet('data/pdbbind_complex.parquet')
requirements.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ mpi4py
2
+ rdkit
3
+ openbabel