-
Notifications
You must be signed in to change notification settings - Fork 39
/
utils.py
514 lines (437 loc) · 12.9 KB
/
utils.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
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
#!/usr/bin/env python
"""
Various sequence manipulation util functions used by different
parts of the pipeline
"""
from __future__ import print_function
try:
from itertools import izip, takewhile
except ImportError:
from itertools import takewhile
izip = zip
import os
import sys
import socket
import pandas as pd
import numpy as np
import string
import ipyrad
BADCHARS = (
string.punctuation
.replace("_", "")
.replace("-", "")
.replace(".", "") + " "
)
class IPyradError(Exception):
"""
Exception handler that does clean exit for CLI, but also prints
the traceback and cleaner message for API.
"""
def __init__(self, *args, **kwargs):
# raise the exception with this string message and a traceback
Exception.__init__(self, *args, **kwargs)
# but suppress traceback and exit on CLI
if not ipyrad.__interactive__:
# clean exit for CLI that still exits as an Error (e.g. for HPC)
sys.tracebacklimit = 0
SystemExit(1)
## utility functions/classes
class Params(object):
"""
A dict-like object for storing params values with a custom repr
that shortens file paths, and which makes attributes easily viewable
through tab completion in a notebook while hiding other funcs, attrs, that
are in normal dicts.
"""
def __len__(self):
return len(self.__dict__)
def __iter__(self):
for attr, value in self.__dict__.items():
yield attr, value
def __getitem__(self, key):
return self.__dict__[key]
def __setitem__(self, key, value):
self.__dict__[key] = value
def __repr__(self):
_repr = ""
keys = sorted(self.__dict__.keys())
if keys:
_printstr = "{:<" + str(2 + max([len(i) for i in keys])) + "} {:<20}\n"
for key in keys:
_val = str(self[key]).replace(os.path.expanduser("~"), "~")
_repr += _printstr.format(key, _val)
return _repr
class ObjDict(dict):
"""
Object dictionary allows calling dictionaries in a more
pretty and Python fashion for storing Assembly data
"""
def __getattr__(self, name):
if name in self:
return self[name]
else:
raise AttributeError("No such attribute: " + name)
def __setattr__(self, name, value):
self[name] = value
def __delattr__(self, name):
if name in self:
del self[name]
else:
raise AttributeError("No such attribute: " + name)
def __repr__(self):
result = ""
if "outfiles" in self.keys():
dirs_order = ["fastqs", "edits", "clusts", "consens", "outfiles"]
for key in dirs_order:
result += key + " : " + self[key] + "\n"
else:
for key in sorted(self):
result += key + " : " + str(self[key]) + "\n"
return result
CDICT = {i: j for i, j in zip("CATG", "0123")}
# ## used for geno output
# VIEW = {
# "R": ("G", "A"),
# "K": ("G", "T"),
# "S": ("G", "C"),
# "Y": ("T", "C"),
# "W": ("T", "A"),
# "M": ("C", "A"),
# "A": ("X", "X"),
# "T": ("X", "X"),
# "G": ("X", "X"),
# "C": ("X", "X"),
# "N": ("X", "X"),
# "-": ("X", "X"),
# }
# ## used in hetero() func of consens_se.py
# TRANS = {
# ('G', 'A'): "R",
# ('G', 'T'): "K",
# ('G', 'C'): "S",
# ('T', 'C'): "Y",
# ('T', 'A'): "W",
# ('C', 'A'): "M",
# }
# # used in write_outfiles.write_geno
TRANSFULL = {
('G', 'A'): "R",
('G', 'T'): "K",
('G', 'C'): "S",
('T', 'C'): "Y",
('T', 'A'): "W",
('C', 'A'): "M",
('A', 'C'): "M",
('A', 'T'): "W",
('C', 'T'): "Y",
('C', 'G'): "S",
('T', 'G'): "K",
('A', 'G'): "R",
}
# TRANSINT = {
# (71, 65): 82,
# (71, 84): 75,
# (71, 67): 83,
# (84, 67): 89,
# (84, 65): 87,
# (67, 65): 77,
# (65, 67): 77,
# (65, 84): 87,
# (67, 84): 89,
# (67, 71): 83,
# (84, 71): 75,
# (65, 71): 82,
# }
# GENOOPTS = {
# b"C": (b"S", b"Y", b"M"),
# b"A": (b"R", b"W", b"M"),
# b"T": (b"K", b"Y", b"W"),
# b"G": (b"R", b"K", b"S"),
# }
## used for resolving ambiguities
AMBIGS = {
"R": ("G", "A"),
"K": ("G", "T"),
"S": ("G", "C"),
"Y": ("T", "C"),
"W": ("T", "A"),
"M": ("C", "A"),
}
def chroms2ints(data, intkeys):
"""
Parse .fai to get a dict with {chroms/scaffolds: ints}, or reversed.
"""
fai = pd.read_csv(
data.params.reference_sequence + ".fai",
names=['scaffold', 'length', 'start', 'a', 'b'],
sep="\t",
)
# Allow CHROM to take integer values, here cast them to str
fai["scaffold"] = fai["scaffold"].astype(str)
faidict = {j: i for i, j in enumerate(fai.scaffold)}
if intkeys:
revdict = {j: i for i, j in faidict.items()}
return revdict
return faidict
def ambigcutters(seq):
"""
Returns both resolutions of a cut site that has an ambiguous base in
it, else the single cut site
"""
resos = []
if any([i in list("RKSYWM") for i in seq]):
for base in list("RKSYWM"):
if base in seq:
resos.append(seq.replace(base, AMBIGS[base][0]))
resos.append(seq.replace(base, AMBIGS[base][1]))
return resos
else:
return [seq, ""]
def splitalleles(consensus):
"""
Takes diploid consensus alleles with phase data stored as a mixture
of upper and lower case characters and splits it into 2 alleles
"""
## store two alleles, allele1 will start with bigbase
allele1 = list(consensus)
allele2 = list(consensus)
hidx = [i for (i, j) in enumerate(consensus) if j in "RKSWYMrkswym"]
## do remaining h sites
for idx in hidx:
hsite = consensus[idx].encode()
if hsite.isupper():
allele1[idx] = PRIORITY[hsite].decode()
allele2[idx] = MINOR[hsite].decode()
else:
allele1[idx] = MINOR[hsite.upper()].decode()
allele2[idx] = PRIORITY[hsite.upper()].decode()
## convert back to strings
allele1 = "".join(allele1)
allele2 = "".join(allele2)
return allele1, allele2
# used by clustmap
def comp(seq):
""" returns a seq with complement. Preserves little n's for splitters."""
## makes base to its small complement then makes upper
return seq.replace("A", 't')\
.replace('T', 'a')\
.replace('C', 'g')\
.replace('G', 'c')\
.replace('n', 'Z')\
.upper()\
.replace("Z", "n")
# used by clustmap
def bcomp(seq):
""" returns a seq with complement. Preserves little n's for splitters."""
## makes base to its small complement then makes upper
return seq.replace(b"A", b't')\
.replace(b'T', b'a')\
.replace(b'C', b'g')\
.replace(b'G', b'c')\
.replace(b'n', b'Z')\
.upper()\
.replace(b"Z", b"n")
# used by rawedit
def fullcomp(seq):
""" returns complement of sequence including ambiguity characters,
and saves lower case info for multiple hetero sequences"""
## this is surely not the most efficient...
seq = seq.replace("A", 'u')\
.replace('T', 'v')\
.replace('C', 'p')\
.replace('G', 'z')\
.replace('u', 'T')\
.replace('v', 'A')\
.replace('p', 'G')\
.replace('z', 'C')
## No complement for S & W b/c complements are S & W, respectively
seq = seq.replace('R', 'u')\
.replace('K', 'v')\
.replace('Y', 'b')\
.replace('M', 'o')\
.replace('u', 'Y')\
.replace('v', 'M')\
.replace('b', 'R')\
.replace('o', 'K')
seq = seq.replace('r', 'u')\
.replace('k', 'v')\
.replace('y', 'b')\
.replace('m', 'o')\
.replace('u', 'y')\
.replace('v', 'm')\
.replace('b', 'r')\
.replace('o', 'k')
return seq
# used by consens
## Alleles priority dict. The key:vals are the same as the AMBIGS dict
## except it returns just one base, w/ the order/priority being (C>A>T>G)
## This dict is used to impute lower case into consens to retain allele
## order for phase in diploids
PRIORITY = {
b"M": b"C",
b"Y": b"C",
b"S": b"C",
b"W": b"A",
b"R": b"A",
b"K": b"T",
}
# The inverse of priority
MINOR = {
b"M": b"A",
b"Y": b"T",
b"S": b"G",
b"W": b"T",
b"R": b"G",
b"K": b"G",
}
# used by write_outputs
# convert byte to list of alleles as ASCII strings
BTS = {
b"R": ["G", "A"],
b"K": ["G", "T"],
b"S": ["G", "C"],
b"Y": ["T", "C"],
b"W": ["T", "A"],
b"M": ["C", "A"],
b"A": ["A", "A"],
b"T": ["T", "T"],
b"G": ["G", "G"],
b"C": ["C", "C"],
b"N": ["N", "N"],
b"-": ["-", "-"]
}
DUCT = {
"R": ["G", "A"],
"K": ["G", "T"],
"S": ["G", "C"],
"Y": ["T", "C"],
"W": ["T", "A"],
"M": ["C", "A"],
"A": ["A", "A"],
"T": ["T", "T"],
"G": ["G", "G"],
"C": ["C", "C"],
"N": ["N", "N"],
"-": ["-", "-"]
}
# GETGENO = np.array([
# list(b"RGA"),
# list(b"KGT"),
# list(b"SGC"),
# list(b"YTC"),
# list(b"WTA"),
# list(b"MCA"),
# list(b"AAA"),
# list(b"TTT"),
# list(b"CCC"),
# list(b"GGG"),
# ], dtype=np.uint8)
# used in baba.py / write_outfiles..py
## with N and - masked to 9
GETCONS = np.array([
[82, 71, 65],
[75, 71, 84],
[83, 71, 67],
[89, 84, 67],
[87, 84, 65],
[77, 67, 65],
[78, 9, 9],
[45, 9, 9],
], dtype=np.uint8)
DCONS = {
82: (71, 65),
75: (71, 84),
83: (71, 67),
89: (84, 67),
87: (84, 65),
77: (67, 65),
78: (9, 9),
45: (9, 9),
67: (67, 67),
65: (65, 65),
84: (84, 84),
71: (71, 71),
}
def clustdealer(pairdealer, optim):
""" return optim clusters given iterators, and whether it got all or not"""
ccnt = 0
chunk = []
while ccnt < optim:
## try refreshing taker, else quit
try:
taker = takewhile(lambda x: x[0] != b"//\n", pairdealer)
oneclust = [b"".join(next(taker))]
except StopIteration:
return 1, chunk
## load one cluster
while 1:
try:
oneclust.append(b"".join(next(taker)))
except StopIteration:
break
chunk.append(b"".join(oneclust))
ccnt += 1
return 0, chunk
def get_threaded_view(ipyclient, split=True):
""" gets optimum threaded view of ids given the host setup """
## engine ids
## e.g., [0, 1, 2, 3, 4, 5, 6, 7, 8]
eids = ipyclient.ids
## get host names
## e.g., ['a', 'a', 'b', 'b', 'a', 'c', 'c', 'c', 'c']
dview = ipyclient.direct_view()
hosts = dview.apply_sync(socket.gethostname)
## group ids into a dict by their hostnames
## e.g., {a: [0, 1, 4], b: [2, 3], c: [5, 6, 7, 8]}
hostdict = {i: [] for i in hosts} # defaultdict(list)
for host, eid in zip(hosts, eids):
hostdict[host].append(eid)
## Now split threads on the same host into separate proc if there are many
hostdictkeys = list(hostdict.keys())
for key in hostdictkeys:
gids = hostdict[key]
maxt = len(gids)
if len(gids) >= 4:
maxt = 2
## if 4 nodes and 4 ppn, put one sample per host
if (len(gids) == 4) and (len(hosts) >= 4):
maxt = 4
if len(gids) >= 6:
maxt = 3
if len(gids) >= 8:
maxt = 4
if len(gids) >= 16:
maxt = 4
## split ids into groups of maxt
threaded = [gids[i:i + maxt] for i in range(0, len(gids), maxt)]
lth = len(threaded)
## if anything was split (lth>1) update hostdict with new proc
if lth > 1:
hostdict.pop(key)
for hostid in range(lth):
hostdict[str(key) + "_" + str(hostid)] = threaded[hostid]
## make sure split numbering is correct
#threaded = hostdict.values()
#assert len(ipyclient.ids) <= len(list(itertools.chain(*threaded)))
return hostdict
##############################################################
def detect_cpus():
"""
Detects the number of CPUs on a system. This is better than asking
ipyparallel since ipp has to wait for Engines to spin up.
"""
# Linux, Unix and MacOS:
if hasattr(os, "sysconf"):
if os.sysconf_names.get("SC_NPROCESSORS_ONLN"):
# Linux & Unix:
ncpus = os.sysconf("SC_NPROCESSORS_ONLN")
if isinstance(ncpus, int) and ncpus > 0:
return ncpus
else: # OSX:
return int(os.popen2("sysctl -n hw.ncpu")[1].read())
# Windows:
if os.environ.get("NUMBER_OF_PROCESSORS"):
ncpus = int(os.environ["NUMBER_OF_PROCESSORS"])
if ncpus > 0:
return ncpus
return 1 # Default