-
Notifications
You must be signed in to change notification settings - Fork 4
/
calculate_bsa.py
executable file
·359 lines (275 loc) · 13.5 KB
/
calculate_bsa.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
import os, sys
import subprocess
from itertools import izip
import gzip
import tempfile
import re
import shlex
import json
import pandas as pd
import numpy as np
import dask.dataframe as dd
from joblib import Parallel, delayed
import freesasa
from Bio.PDB import PDBList
from util import data_path_prefix, get_interfaces_path, iter_cdd
NUM_WORKERS = 4
RAW_PDB_PATH = os.path.join(data_path_prefix, "pdb", "pdb")
PDB_TOOLS = os.path.join(os.path.dirname(data_path_prefix), "pdb-tools")
cutoffs = {
"weak transient": (0, 1500),
"transient": (1500, 2500),
"permanent": (2500, float("inf"))
}
def get_pdb(pdb, chain1, chain2, sdi1, sdi2):
prefix = "freesasa_{}_{}_{}_full.pdb".format(pdb, chain1, chain2)
tmp_pdb = tempfile.NamedTemporaryFile(prefix=prefix, delete=False)
tmp_pdb_path = tmp_pdb.name
pdb_path = os.path.join(RAW_PDB_PATH, pdb[1:3].lower(), "pdb{}.ent.gz".format(pdb.lower()))
if os.path.isfile(pdb_path):
with tmp_pdb as tmp, gzip.open(pdb_path, 'rb') as pdbf:
for line in pdbf:
tmp.write(line)
else:
PDBList().retrieve_pdb_file(pdb, obsolete=True, pdir=os.getcwd(), file_format="pdb")
tmp_pdb_path = "pdb{}.ent".format(pdb.lower())
if not os.path.isfile(tmp_pdb_path) or os.path.getsize(tmp_pdb_path) == 0:
raise IOError("Cannot load "+pdb)
tmp_pdb.close()
prefix = "freesasa_{}_{}_{}_d1d2.pdb".format(pdb, chain1, chain2)
tmp_pdb2 = tempfile.NamedTemporaryFile(prefix=prefix, delete=False)
with open(tmp_pdb2.name, "w") as output:
splitchain1 = subprocess.Popen(
[sys.executable, os.path.join(PDB_TOOLS, "pdb_selchain.py"), "-{}".format(chain1), tmp_pdb_path],
stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
splitdomain1 = subprocess.Popen(
[sys.executable, os.path.join(PDB_TOOLS, "pdb_rslice.py")]+list(sdi1),
stdin=splitchain1.stdout,
stdout=output,
stderr=subprocess.PIPE)
out1, err1 = splitdomain1.communicate()
if chain2 is not None and sdi2 is not None:
splitchain2 = subprocess.Popen(
[sys.executable, os.path.join(PDB_TOOLS, "pdb_selchain.py"), "-{}".format(chain2), tmp_pdb_path],
stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
splitdomain2 = subprocess.Popen(
[sys.executable, os.path.join(PDB_TOOLS, "pdb_rslice.py")]+list(sdi2),
stdin=splitchain2.stdout,
stdout=output,
stderr=subprocess.PIPE)
out2, err2 = splitdomain2.communicate()
print "ERR1", out1, err1, tmp_pdb.name, sdi1
#os.remove(tmp_pdb.name)
return tmp_pdb2.name
def run_freesasa(command):
print " ".join(command)
FNULL = open(os.devnull, 'w')
freesasa = subprocess.check_output(command, stderr=FNULL)
freesasa = "{"+freesasa.split("\n",1)[1]
return json.loads(freesasa)
def calculate_buried_surface_area(pdb_file, pdb, chain1, chain2, sdi_sel1=None, sdi_sel2=None, face1=None, face2=None):
chains = [chain1+chain2, chain1, chain2]
command = ["freesasa", "--format=json", "--chain-groups={}".format("+".join(chains))]
if sdi_sel1 is not None and sdi_sel1 is not None:
command.append("--select='domain1, chain {} and resi {}'".format(chain1, "+".join(sdi_sel1)))
command.append("--select='domain2, chain {} and resi {}'".format(chain2, "+".join(sdi_sel2)))
if face1 is not None:
residues = face1.replace(",", "+")
selection = "binding-site1, chain {} and resi {}".format(chain1, residues)
command.append("--select='{}'".format(selection))
if face2 is not None:
residues = face2.replace(",", "+")
selection = "binding-site2, chain {} and resi {}".format(chain2, residues)
command.append("--select='{}'".format(selection))
command.append(pdb_file)
freesasa = run_freesasa(command)
if sdi_sel1 is not None and sdi_sel1 is not None:
try:
c1_asa = freesasa["results"][2]["structure"][0]["selections"][0]["area"]
except (IndexError, KeyError):
#Use full chain area, it is assumed the interacting domains are split out
c1_asa = freesasa["results"][2]["structure"][0]["area"]["total"]
try:
c2_asa = freesasa["results"][3]["structure"][0]["selections"][0]["area"]
except (IndexError, KeyError):
#Use full chain area, it is assumed the interacting domains are split out
c2_asa = freesasa["results"][3]["structure"][0]["area"]["total"]
complex_asa = sum([sel["area"] for sel in freesasa["results"][1]["structure"][0]["selections"] \
if "domain" in sel["name"]])
print complex_asa, freesasa["results"][1]["structure"][0]["area"]["total"]
if complex_asa == 0.0:
complex_asa = freesasa["results"][1]["structure"][0]["area"]["total"]
print "Error, used total", complex_asa, freesasa["results"][1]["structure"][0]["selections"], command
else:
c1_asa = freesasa["results"][2]["structure"][0]["area"]["total"]
c2_asa = freesasa["results"][3]["structure"][0]["area"]["total"]
complex_asa = freesasa["results"][1]["structure"][0]["area"]["total"]
bsa = c1_asa+c2_asa-complex_asa
for ppi_type, (low_cut, high_cut) in cutoffs.iteritems():
if low_cut <= bsa < high_cut:
return bsa, ppi_type, c1_asa, c2_asa, complex_asa
else:
return bsa, "unknown", c1_asa, c2_asa, complex_asa
def calculate_surface_area_chain(pdb_file, pdb, chain, sdi=None, face=None):
command = ["freesasa", "--format=json", "--chain-groups={}".format(chain)]
if sdi is not None:
command.append("--select=domain, resi {}".format(sdi))
if face is not None:
residues = face.replace(",", "+")
selection = "binding-site, chain {} and resi {}".format(chain, residues)
command.append("--select='{}'".format(selection))
command.append(pdb_file)
freesasa = run_freesasa(command)
if sdi is not None:
try:
return freesasa["results"][1]["structure"][0]["selections"][0]["area"]
except (IndexError, KeyError):
pass
return freesasa["results"][1]["structure"][0]["area"]["total"]
def get_bsa(df):
print "DF is {} {} {} {}".format(type(df), df.shape, df.iloc[0].mol_sdi, df)
r = df.iloc[0]
if any(r[["mol_chain", "int_chain"]].isna()):
#Cannot process chain, so bsa is unknown
return pd.Series({"mol_sdi":np.nan, "bsa":np.nan, "ppi_type":"unknown", "c1_asa":np.nan, "c2_asa":np.nan, "complex_asa":np.nan})
#If interacting domain is not well defined, use the entire chain
df[["int_sdi_from"]] = df[["int_sdi_from"]].fillna(1)
df[["int_sdi_to"]] = df[["int_sdi_to"]].fillna(0)
try:
sdi1, sdi2 = zip(*[(
"{}:{}".format(int(row.mol_sdi_from), int(row.mol_sdi_to)),
"{}:{}".format(int(row.int_sdi_from), int(row.int_sdi_to) if row.int_sdi_to else "")) \
for row in df.itertuples()])
except ValueError:
#There is a NaN in mol sdi or from to: Invalid!
return pd.Series({"mol_sdi":np.nan, "bsa":np.nan, "ppi_type":"unknown", "c1_asa":np.nan, "c2_asa":np.nan, "complex_asa":np.nan})
try:
pdb_file = get_pdb(r["mol_pdb"], r["mol_chain"], r["int_chain"], sdi1, sdi2)
except IOError:
return pd.Series({"mol_sdi":np.nan, "bsa":np.nan, "ppi_type":"unknown", "c1_asa":np.nan, "c2_asa":np.nan, "complex_asa":np.nan})
bsa, ppi_type, c1_asa, c2_asa, complex_asa = calculate_buried_surface_area(
pdb_file, r["mol_pdb"], r["mol_chain"], r["int_chain"],
face1=r["mol_res"], face2=r["int_res"])
try:
os.remove(pdb_file)
except OSError:
pass
return pd.Series({"mol_sdi":r.mol_sdi, "bsa":bsa, "ppi_type":ppi_type, "c1_asa":c1_asa, "c2_asa":c2_asa, "complex_asa":complex_asa})
def observed_bsa(job, dataset_name, cdd, cores=NUM_WORKERS):
job.log("CDD {}".format(cdd))
prefix = os.path.join(get_interfaces_path(dataset_name), "by_superfamily", str(int(cdd)), str(int(cdd)))
if os.path.isfile(prefix+"_bsa.h5"):
store = pd.HDFStore(unicode(prefix+"_bsa.h5"))
if "/observed" in store.keys():
store.close()
return
store.close()
cdd_interactome_path = prefix+".observed_interactome"
print cdd_interactome_path
cdd_interactome = pd.read_hdf(unicode(cdd_interactome_path), "table")
if cdd_interactome.shape[0] == 0:
job.log("CDD observed interactome is empty -- FIX!!!")
return
job.log("KEYS: {}".format(cdd_interactome.columns))
job.log("{}".format(cdd_interactome))
cdd_interactome = cdd_interactome[cdd_interactome["mol_chain"] != cdd_interactome["int_chain"]]
if cdd_interactome.shape[0] == 0:
job.log("CDD observed interactome contains intra-chain PPI, skipped -- FIX!!!")
return
#Remove redundant interfaces
cdd_interactome = cdd_interactome.groupby(["obs_int_id", "mol_sdi_from", "mol_sdi_to"],
as_index=False).nth(0).reset_index(drop=True).copy()
print cdd_interactome
job.log("KEYS: {} {}".format(cdd_interactome.columns, cdd_interactome.shape))
bsa = Parallel(n_jobs=NUM_WORKERS)(delayed(get_bsa)(group) for _, group in \
cdd_interactome.groupby("mol_sdi", as_index=False))
bsa = pd.DataFrame(bsa)
assert bsa.shape[0] > 0, "{} {}".format(cdd, bsa)
cdd_interactome = pd.merge(cdd_interactome, bsa, how="left", on="mol_sdi")
# df = dd.from_pandas(cdd_interactome, npartitions=NUM_WORKERS)
#
# meta = pd.DataFrame({"bsa":[1.], "ppi_type":["0"], "c1_asa":[1.], "c2_asa":[1.], "complex_asa":[1.]})
# bsa = df.map_partitions(lambda _df: _df.apply(get_bsa, axis=1), meta=meta)\
# .compute(scheduler="multiprocessing", num_workers=NUM_WORKERS)
# cdd_interactome[bsa.columns] = bsa
cdd_interactome.to_hdf(unicode(prefix+"_bsa.h5"), "observed", complevel=9, complib="bzip2")
def get_asa(df):
df = df.reset_index(drop=True)
r = df.iloc[0]
try:
sdi1 = ["{}:{}".format(int(row.mol_sdi_from), int(row.mol_sdi_to)) for row in df.itertuples()]
except:
print "Failed due to being Series? {}".format(type(df))
print "DF is {}".format(df)
raise
try:
pdb_file = get_pdb(r["mol_pdb"], r["mol_chain"], None, sdi1, None)
except IOError:
return pd.Series({"mol_sdi":np.nan, "c2_asa_pred":np.nan, "pred_ratio":np.nan})
asa = calculate_surface_area_chain(
pdb_file, r.mol_pdb, r.mol_chain, face=r.mol_resi)
try:
os.remove(pdb_file)
except OSError:
pass
predicted_bsa = asa+r["c2_asa"]-r["complex_asa"]
ratio = predicted_bsa/r["bsa"]
for ppi_type, (low_cut, high_cut) in cutoffs.iteritems():
if low_cut <= predicted_bsa < high_cut:
break
else:
ppi_type = "unknown"
return pd.Series({"mol_sdi":df.mol_sdi, "c2_asa_pred":predicted_bsa, "pred_ratio":ratio, "ppi_type_pred":ppi_type})
def inferred_bsa(job, dataset_name, cdd, cores=NUM_WORKERS):
job.log("INF CDD {}".format(cdd))
cdd_bsa_path = os.path.join(get_interfaces_path(dataset_name), "by_superfamily", str(int(cdd)), str(int(cdd)))
if not os.path.isfile(cdd_bsa_path+"_bsa.h5"):
job.log("observed bsa must exist")
return
store = pd.HDFStore(unicode(cdd_bsa_path+"_bsa.h5"))
if "/inferred" in store.keys():
return
try:
cdd_obs_bsa = store.get("/observed")
except KeyError:
raise RuntimeError("Must calculate observed BSAs first")
try:
cdd_obs_bsa = cdd_obs_bsa[["obs_int_id", "bsa", "c1_asa", "c2_asa", "complex_asa", "ppi_type"]]
except KeyError:
job.log("Failed due to column select {}".format(cdd_obs_bsa.columns))
raise
inf_interactome = pd.read_hdf(unicode(cdd_bsa_path+".inferred_interactome"), "table")
inf_interactome = pd.merge(inf_interactome, cdd_obs_bsa, how="left", left_on="nbr_obs_int_id", right_on="obs_int_id")
del inf_interactome["obs_int_id"]
#Remove redundant interfaces
inf_interactome = inf_interactome.groupby(["mol_sdi", "nbr_obs_int_id", "mol_sdi_from", "mol_sdi_to"],
as_index=False).nth(0).reset_index(drop=True).copy()
bsa = Parallel(n_jobs=NUM_WORKERS)(delayed(get_asa)(group) for _, group in \
inf_interactome.groupby(["mol_sdi", "nbr_obs_int_id"], as_index=False))
bsa = pd.DataFrame(bsa)
bsa["mol_sdi"] = bsa["mol_sdi"].astype(float)
inf_interactome = pd.merge(inf_interactome, bsa, how="left", on="mol_sdi")
inf_interactome.to_hdf(unicode(cdd_bsa_path+"_bsa.h5"), "inferred", complevel=9, complib="bzip2")
def start_toil(job, dataset_name, name="bsa"):
path = os.path.join(get_interfaces_path(dataset_name), "by_superfamily")
for cdd, sfam_id in iter_cdd(use_id=True, group_superfam=True):
sfam_path = os.path.join(path, str(int(sfam_id)), str(int(sfam_id)))
if not os.path.isfile(sfam_path+".observed_interactome"):
continue
cjob = job.addChildJobFn(observed_bsa, dataset_name, sfam_id)
if not os.path.isfile(sfam_path+".inferred_interactome"):
continue
cjob.addFollowOnJobFn(inferred_bsa, dataset_name, sfam_id)
if __name__ == "__main__":
from toil.common import Toil
from toil.job import Job
parser = Job.Runner.getDefaultArgumentParser()
options = parser.parse_args()
options.logLevel = "DEBUG"
options.clean = "always"
dataset_name = options.jobStore.split(":")[-1]
job = Job.wrapJobFn(start_toil, dataset_name)
with Toil(options) as toil:
toil.start(job)