-
Notifications
You must be signed in to change notification settings - Fork 39
/
baba.py
1215 lines (976 loc) · 39.8 KB
/
baba.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
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
#!/usr/bin/env python
""" D-statistic calculations """
# pylint: disable=E1101
# pylint: disable=F0401
# pylint: disable=W0142
# pylint: disable=R0915
# pylint: disable=R0914
# pylint: disable=R0912
from __future__ import print_function, division
from ipyrad.assemble.write_outputs import reftrick
from ipyrad.assemble.utils import IPyradError, GETCONS, Params
from ipyrad.analysis.utils import progressbar
# from ipyrad.plotting.baba_panel_plot import baba_panel_plot
# import scipy.stats as st ## used for dfoil
import pandas as pd
import numpy as np
import numba
import itertools
import datetime
import types
import copy
import time
import sys
import os
# ipyrad tools
from ipyrad.analysis.utils import Params, progressbar
from ipyrad.assemble.utils import IPyradError
from ipyrad.assemble.write_outputs import reftrick
try:
import toytree
except ImportError:
pass
_TOYTREE_IMPORT = """
This ipyrad analysis tool requires
You can install it with the following command:
conda install toytree -c eaton-lab
"""
# try:
# import toyplot
# except ImportError:
# pass
# _TOYPLOT_IMPORT = """
# This ipyrad analysis tool requires the toyplot package.
# You can install it with the following command:
# conda install toyplot -c eaton-lab
# """
# set floating point precision in data frames to 3 for prettier printing
pd.set_option('precision', 3)
class Baba(object):
"new baba class object"
def __init__(
self,
data=None,
tests=None,
newick=None,
nboots=1000,
mincov=1):
"""
ipyrad.analysis Baba Class object.
Parameters
----------
data : string or ndarray
A string path to a .loci file produced by ipyrad. Alternatively,
data can be entered as a Numpy array of float allele frequencies
with dimension (nloci, 4 or 5, maxlen). See simulation example
in the docs.
tests : dict or list of dicts
A dictionary mapping Sample names to test taxon names, e.g.,
test = {'p1': ['a', 'b'], 'p2': ['c'], 'p3': ['e'], 'p4': ['f']}.
Four taxon tests should have p1-p4 whereas five taxon tests will
used if dict keys are p1-p5. Other key names will raise an error.
The highest value name (e.g., p5) is the outgroup.
newick: str
...
Functions
---------
run()
...
generate_tests_from_tree()
...
plot()
...
"""
# check external imports
if not sys.modules.get("toytree"):
raise ImportError(_TOYTREE_IMPORT)
if not sys.modules.get("toyplot"):
raise ImportError(_TOYPLOT_IMPORT)
## parse data as (1) path to data file, or (2) ndarray
if isinstance(data, str):
self.data = os.path.realpath(data)
else:
self.data = data
if isinstance(newick, toytree.tree):
self.newick = newick.tree.write()
else:
self.newick = newick
## store tests, check for errors
self.tests = tests
## parameters
self.params = Params()
self.params.mincov = mincov
self.params.nboots = nboots
self.params.quiet = False
self.params.database = None
## results storage
self.results_table = None
self.results_boots = None
@property
def taxon_table(self):
"""
Returns the .tests list of taxa as a pandas dataframe.
By auto-generating this table from tests it means that
the table itself cannot be modified unless it is returned
and saved.
"""
if self.tests:
keys = sorted(self.tests[0].keys())
if isinstance(self.tests, list):
ld = [[(key, i[key]) for key in keys] for i in self.tests]
dd = [dict(i) for i in ld]
df = pd.DataFrame(dd)
return df
else:
return pd.DataFrame(pd.Series(self.tests)).T
else:
return None
def run(self,
ipyclient=None,
):
"""
Run a batch of dstat tests on a list of tests, where each test is
a dictionary mapping sample names to {p1 - p4} (and sometimes p5).
Parameters modifying the behavior of the run, such as the number
of bootstrap replicates (nboots) or the minimum coverage for
loci (mincov) can be set in {object}.params.
Parameters:
-----------
ipyclient (ipyparallel.Client object):
An ipyparallel client object to distribute jobs to a cluster.
"""
self.results_table, self.results_boots = batch(self, ipyclient)
## skip this for 5-part test results
if not isinstance(self.results_table, list):
self.results_table.nloci = np.nan_to_num(self.results_table.nloci)\
.astype(int)
def generate_tests_from_tree(self,
constraint_dict=None,
constraint_exact=False,
verbose=True):
"""
Returns a list of all possible 4-taxon tests on a tree (newick file).
The number of possible tests can be greatly reduced by setting
constraints on the taxon sampling using the constraint_dict arg.
Parameters:
-----------
constraint_dict: dict
The constraint dict will limit the tests generated to only include
the taxa listed in the dict.
constraint_exact: bool or list
If constraint_exact is True then only samples meeting the exact
entries in the constraint_dict will be returned, as opposed to all
subsets of those entries. If a list then different values can be
applied to [p1, p2, p3, p4]. For example, if the constraint_dict is
{"p1": sample1, "p2": sample2, "p3": sample3, "p4": [sample4, sample5]},
then with constraint_exact==False you get:
sample1, sample2, sample3, sample4
sample1, sample2, sample3, sample5
sample1, sample2, sample3, [sample4, sample5]
and with constraint_exact==True you get only:
sample1, sample2, sample3, [sample4, sample5]
"""
if not self.newick:
raise AttributeError("no newick tree information in {self}.newick")
tests = tree2tests(self.newick, constraint_dict, constraint_exact)
if verbose:
print("{} tests generated from tree".format(len(tests)))
self.tests = tests
def plot(self,
show_test_labels=True,
use_edge_lengths=True,
collapse_outgroup=False,
pct_tree_x=0.5,
pct_tree_y=0.2,
subset_tests=None,
#toytree_kwargs=None,
*args,
**kwargs):
"""
Draw a multi-panel figure with tree, tests, and results
Parameters:
-----------
height: int
...
width: int
...
show_test_labels: bool
...
use_edge_lengths: bool
...
collapse_outgroups: bool
...
pct_tree_x: float
...
pct_tree_y: float
...
subset_tests: list
...
...
"""
print("Plotting baba results is not implemented in v.0.9.")
return
## check for attributes
if not self.newick:
raise IPyradError("baba plot requires a newick treefile")
if not self.tests:
raise IPyradError("baba plot must have a .tests attribute")
## ensure tests is a list
if isinstance(self.tests, dict):
self.tests = [self.tests]
## re-decompose the tree
ttree = toytree.tree(
self.newick,
orient='down',
use_edge_lengths=use_edge_lengths,
)
## subset test to show fewer
if subset_tests != None:
#tests = self.tests[subset_tests]
tests = [self.tests[i] for i in subset_tests]
boots = self.results_boots[subset_tests]
else:
tests = self.tests
boots = self.results_boots
## make the plot
#canvas, axes, panel = baba_panel_plot(
# ttree=ttree,
# tests=tests,
# boots=boots,
# show_test_labels=show_test_labels,
# use_edge_lengths=use_edge_lengths,
# collapse_outgroup=collapse_outgroup,
# pct_tree_x=pct_tree_x,
# pct_tree_y=pct_tree_y,
# *args,
# **kwargs)
#return canvas, axes, panel
def copy(self):
""" returns a copy of the baba analysis object """
return copy.deepcopy(self)
def batch(
baba,
ipyclient=None,
):
"""
distributes jobs to the parallel client
"""
## parse args
handle = baba.data
taxdicts = baba.tests
mindicts = baba.params.mincov
nboots = baba.params.nboots
## if ms generator make into reusable list
sims = 0
if isinstance(handle, types.GeneratorType):
handle = list(handle)
sims = 1
else:
## expand locifile path to full path
handle = os.path.realpath(handle)
## parse taxdicts into names and lists if it a dictionary
#if isinstance(taxdicts, dict):
# names, taxdicts = taxdicts.keys(), taxdicts.values()
#else:
# names = []
names = []
if isinstance(taxdicts, dict):
taxdicts = [taxdicts]
## an array to hold results (len(taxdicts), nboots)
tot = len(taxdicts)
resarr = np.zeros((tot, 7), dtype=np.float64)
bootsarr = np.zeros((tot, nboots), dtype=np.float64)
paneldict = {}
## TODO: Setup a wrapper to find and cleanup ipyclient
## define the function and parallelization to use,
## if no ipyclient then drops back to using multiprocessing.
if not ipyclient:
# ipyclient = ip.core.parallel.get_client(**self._ipcluster)
raise IPyradError("you must enter an ipyparallel.Client() object")
else:
lbview = ipyclient.load_balanced_view()
## submit jobs to run on the cluster queue
start = time.time()
asyncs = {}
idx = 0
## prepare data before sending to engines
## if it's a str (locifile) then parse it here just once.
if isinstance(handle, str):
with open(handle, 'r') as infile:
loci = infile.read().strip().split("|\n")
if isinstance(handle, list):
pass #sims()
## iterate over tests (repeats mindicts if fewer than taxdicts)
itests = iter(taxdicts)
imdict = itertools.cycle([mindicts])
#for test, mindict in zip(taxdicts, itertools.cycle([mindicts])):
for i in range(len(ipyclient)):
## next entries unless fewer than len ipyclient, skip
try:
test = next(itests)
mindict = next(imdict)
except StopIteration:
continue
## if it's sim data then convert to an array
if sims:
loci = _msp_to_arr(handle, test)
args = (loci, test, mindict, nboots)
print("not yet implemented")
#asyncs[idx] = lbview.apply_async(dstat, *args)
else:
args = [loci, test, mindict, nboots]
asyncs[idx] = lbview.apply(dstat, *args)
idx += 1
## block until finished, print progress if requested.
finished = 0
try:
while 1:
keys = [i for (i, j) in asyncs.items() if j.ready()]
## check for failures
for job in keys:
if not asyncs[job].successful():
raise IPyradError(\
" error: {}: {}".format(job, asyncs[job].exception()))
## enter results for successful jobs
else:
_res, _bot = asyncs[job].result()
## store D4 results
if _res.shape[0] == 1:
resarr[job] = _res.T.values[:, 0]
bootsarr[job] = _bot
## or store D5 results
else:
paneldict[job] = _res.T
## remove old job
del asyncs[job]
finished += 1
## submit next job if there is one.
try:
test = next(itests)
mindict = next(imdict)
if sims:
loci = _msp_to_arr(handle, test)
args = (loci, test, mindict, nboots)
print("not yet implemented")
#asyncs[idx] = lbview.apply_async(dstat, *args)
else:
args = [loci, test, mindict, nboots]
asyncs[idx] = lbview.apply(dstat, *args)
idx += 1
except StopIteration:
pass
## count finished and break if all are done.
#fin = idx - len(asyncs)
elap = datetime.timedelta(seconds=int(time.time()-start))
printstr = " calculating D-stats"
progressbar(finished, tot, start, message=printstr)
time.sleep(0.1)
if not asyncs:
print("")
break
except KeyboardInterrupt as inst:
## cancel all jobs (ipy & multiproc modes) and then raise error
try:
ipyclient.abort()
except Exception:
pass
raise inst
## dress up resarr as a Pandas DataFrame if 4-part test
if len(test) == 4:
if not names:
names = range(len(taxdicts))
#print("resarr")
#print(resarr)
resarr = pd.DataFrame(resarr,
index=names,
columns=["dstat", "bootmean", "bootstd", "Z", "ABBA", "BABA", "nloci"])
## sort results and bootsarr to match if test names were supplied
resarr = resarr.sort_index()
order = [list(resarr.index).index(i) for i in names]
bootsarr = bootsarr[order]
return resarr, bootsarr
else:
## order results dfs
listres = []
for key in range(len(paneldict)):
listres.append(paneldict[key])
## make into a multi-index dataframe
ntests = len(paneldict)
multi_index = [
np.array([[i] * 3 for i in range(ntests)]).flatten(),
np.array(['p3', 'p4', 'shared'] * ntests),
]
resarr = pd.DataFrame(
data=pd.concat(listres).values,
index=multi_index,
columns=listres[0].columns,
)
return resarr, None
#return listres, None #_res.T, _bot
def dstat(inarr, taxdict, mindict=1, nboots=1000, name=0):
""" private function to perform a single D-stat test"""
#if isinstance(inarr, str):
# with open(inarr, 'r') as infile:
# inarr = infile.read().strip().split("|\n")
# ## get data as an array from loci file
# ## if loci-list then parse arr from loci
if isinstance(inarr, list):
arr, _ = _loci_to_arr(inarr, taxdict, mindict)
# ## if it's an array already then go ahead
# elif isinstance(inarr, np.ndarray):
# arr = inarr
# ## if it's a simulation object get freqs from array
# elif isinstance(inarr, Sim):
# arr = _msp_to_arr(inarr, taxdict)
#elif isinstance(inarr, types.GeneratorType):
# arr = _msp_to_arr(inarr, taxdict)
#elif isinstance(inarr, list):
# arr = _msp_to_arr(inarr, taxdict)
## get data from Sim object, do not digest the ms generator
#else:
# raise Exception("Must enter either a 'locifile' or 'arr'")
## run tests
#if len(taxdict) == 4:
if arr.shape[1] == 4:
## get results
res, boots = _get_signif_4(arr, nboots)
## make res into a nice DataFrame
res = pd.DataFrame(res,
columns=[name],
index=["Dstat", "bootmean", "bootstd", "Z", "ABBA", "BABA", "nloci"])
else:
## get results
res, boots = _get_signif_5(arr, nboots)
## make int a DataFrame
res = pd.DataFrame(res,
index=["p3", "p4", "shared"],
columns=["Dstat", "bootmean", "bootstd", "Z", "ABxxA", "BAxxA", "nloci"]
)
return res.T, boots
def _loci_to_arr(loci, taxdict, mindict):
"""
return a frequency array from a loci file for all loci with taxa from
taxdict and min coverage from mindict.
"""
## make the array (4 or 5) and a mask array to remove loci without cov
nloci = len(loci)
maxlen = np.max(np.array([len(locus.split("\n")[0]) for locus in loci]))
keep = np.zeros(nloci, dtype=np.bool_)
arr = np.zeros((nloci, 4, maxlen), dtype=np.float64)
## six rows b/c one for each p3, and for the fused p3 ancestor
if len(taxdict) == 5:
arr = np.zeros((nloci, 6, maxlen), dtype=np.float64)
## if not mindict, make one that requires 1 in each taxon
if isinstance(mindict, int):
mindict = {i: mindict for i in taxdict}
elif isinstance(mindict, dict):
mindict = {i: mindict[i] for i in taxdict}
else:
mindict = {i: 1 for i in taxdict}
## raise error if names are not 'p[int]'
allowed_names = ['p1', 'p2', 'p3', 'p4', 'p5']
if any([i not in allowed_names for i in taxdict]):
raise IPyradError(\
"keys in taxdict must be named 'p1' through 'p4' or 'p5'")
## parse key names
keys = sorted([i for i in taxdict.keys() if i[0] == 'p'])
outg = keys[-1]
## grab seqs just for the good guys
for loc in range(nloci):
try:
## parse the locus
lines = loci[loc].split("\n")[:-1]
names = [i.split()[0] for i in lines]
seqs = np.array([list(i.split()[1]) for i in lines])
except Exception as inst:
raise IPyradError("Malformed locus\n{}".format(loci))
## check that names cover the taxdict (still need to check by site)
covs = [sum([j in names for j in taxdict[tax]]) >= mindict[tax] \
for tax in taxdict]
## keep locus
if all(covs):
keep[loc] = True
## get the refseq
refidx = np.where([i in taxdict[outg] for i in names])[0]
refseq = seqs[refidx].view(np.uint32)
ancestral = np.array([reftrick(refseq, GETCONS)[:, 0]])
## freq of ref in outgroup
iseq = _reffreq2(ancestral, refseq, GETCONS)
try:
arr[loc, -1, :iseq.shape[1]] = iseq
except:
raise IPyradError("Error in getting outgroup sequences.")
## enter 4-taxon freqs
if len(taxdict) == 4:
for tidx, key in enumerate(keys[:-1]):
## get idx of names in test tax
nidx = np.where([i in taxdict[key] for i in names])[0]
sidx = seqs[nidx].view(np.uint32)
## get freq of sidx
iseq = _reffreq2(ancestral, sidx, GETCONS)
## fill it in
arr[loc, tidx, :iseq.shape[1]] = iseq
else:
## entere p5; and fill it in
iseq = _reffreq2(ancestral, refseq, GETCONS)
arr[loc, -1, :iseq.shape[1]] = iseq
## enter p1
nidx = np.where([i in taxdict['p1'] for i in names])[0]
sidx = seqs[nidx].view(np.uint32)
iseq = _reffreq2(ancestral, sidx, GETCONS)
arr[loc, 0, :iseq.shape[1]] = iseq
## enter p2
nidx = np.where([i in taxdict['p2'] for i in names])[0]
sidx = seqs[nidx].view(np.uint32)
iseq = _reffreq2(ancestral, sidx, GETCONS)
arr[loc, 1, :iseq.shape[1]] = iseq
## enter p3 with p4 masked, and p4 with p3 masked
nidx = np.where([i in taxdict['p3'] for i in names])[0]
nidy = np.where([i in taxdict['p4'] for i in names])[0]
sidx = seqs[nidx].view(np.uint32)
sidy = seqs[nidy].view(np.uint32)
xseq = _reffreq2(ancestral, sidx, GETCONS)
yseq = _reffreq2(ancestral, sidy, GETCONS)
mask3 = xseq != 0
mask4 = yseq != 0
xseq[mask4] = 0
yseq[mask3] = 0
arr[loc, 2, :xseq.shape[1]] = xseq
arr[loc, 3, :yseq.shape[1]] = yseq
## enter p34
nidx = nidx.tolist() + nidy.tolist()
sidx = seqs[nidx].view(np.uint32)
iseq = _reffreq2(ancestral, sidx, GETCONS)
arr[loc, 4, :iseq.shape[1]] = iseq
## size-down array to the number of loci that have taxa for the test
arr = arr[keep, :, :]
## size-down sites to
arr = masknulls(arr)
return arr, keep
## This should be re-written as a dynamic func
def tree2tests(newick, constraint_dict, constraint_exact):
"""
Returns dict of all possible four-taxon splits in a tree. Assumes
the user has entered a rooted tree. Skips polytomies.
"""
## make tree
tree = toytree.tree(newick)
testset = set()
## expand constraint_exact if list
if isinstance(constraint_exact, bool):
constraint_exact = [constraint_exact] * 4
elif isinstance(constraint_exact, list):
if len(constraint_exact) != len(constraint_dict):
raise Exception("constraint_exact must be bool or [bool, bool, bool, bool]")
## constraints
cdict = {"p1":[], "p2":[], "p3":[], "p4":[]}
if constraint_dict:
cdict.update(constraint_dict)
## traverse root to tips. Treat the left as outgroup, then the right.
tests = []
## topnode must have children
for topnode in tree.treenode.traverse("levelorder"):
for oparent in topnode.children:
for onode in oparent.traverse("levelorder"):
if test_constraint(onode, cdict, "p4", constraint_exact[3]):
#print(topnode.name, onode.name)
## p123 parent is sister to oparent
p123parent = oparent.get_sisters()[0]
for p123node in p123parent.traverse("levelorder"):
for p3parent in p123node.children:
for p3node in p3parent.traverse("levelorder"):
if test_constraint(p3node, cdict, "p3", constraint_exact[2]):
#print(topnode.name, onode.name, p3node.name)
## p12 parent is sister to p3parent
p12parent = p3parent.get_sisters()[0]
for p12node in p12parent.traverse("levelorder"):
for p2parent in p12node.children:
for p2node in p2parent.traverse("levelorder"):
if test_constraint(p2node, cdict, "p2", constraint_exact[1]):
## p12 parent is sister to p3parent
p1parent = p2parent.get_sisters()[0]
for p1node in p1parent.traverse("levelorder"):
#for p1parent in p1node.children:
# for p1node in p1parent.traverse("levelorder"):
if test_constraint(p1node, cdict, "p1", constraint_exact[0]):
x = (onode.name, p3node.name, p2node.name, p1node.name)
test = {}
test['p4'] = onode.get_leaf_names()
test['p3'] = p3node.get_leaf_names()
test['p2'] = p2node.get_leaf_names()
test['p1'] = p1node.get_leaf_names()
if x not in testset:
tests.append(test)
testset.add(x)
return tests
def test_constraint(node, cdict, tip, exact):
names = set(node.get_leaf_names())
const = set(cdict[tip])
if const:
if exact:
#if len(names.intersection(const)) == len(const):
if names == const:
return 1
else:
return 0
else:
if len(names.intersection(const)) == len(names):
return 1
else:
return 0
return 1
@numba.jit(nopython=True)
def masknulls(arr):
nvarr = np.zeros(arr.shape[0], dtype=np.int8)
trimarr = np.zeros(arr.shape, dtype=np.float64)
for loc in range(arr.shape[0]):
nvars = 0
for site in range(arr.shape[2]):
col = arr[loc, :, site]
## mask cols with 9s
if not np.any(col == 9):
## any non-outgroup shows variation?
## todo: check whether BBBBA is ever info?
if np.any(col[:-1] != col[0]):
trimarr[loc, :, nvars] = col
nvars += 1
nvarr[loc] = nvars
return trimarr[:, :, :nvarr.max()]
@numba.jit(nopython=True)
def _reffreq2(ancestral, iseq, consdict):
## empty arrays
freq = np.zeros((1, iseq.shape[1]), dtype=np.float64)
amseq = np.zeros((iseq.shape[0]*2, iseq.shape[1]), dtype=np.uint8)
## fill in both copies
for seq in range(iseq.shape[0]):
for col in range(iseq.shape[1]):
## get this base and check if it is hetero
base = iseq[seq][col]
who = consdict[:, 0] == base
## if not hetero then enter it
if not np.any(who):
amseq[seq*2][col] = base
amseq[seq*2+1][col] = base
## if hetero then enter the 2 resolutions
else:
amseq[seq*2][col] = consdict[who, 1][0]
amseq[seq*2+1][col] = consdict[who, 2][0]
## amseq may have N or -, these need to be masked
for i in range(amseq.shape[1]):
## without N or -
reduced = amseq[:, i][amseq[:, i] != 9]
counts = reduced != ancestral[0][i]
if reduced.shape[0]:
freq[:, i] = counts.sum() / reduced.shape[0]
else:
freq[:, i] = 9
return freq
@numba.jit(nopython=True)
def _prop_dstat(arr):
## numerator
abba = ((1.-arr[:, 0]) * (arr[:, 1]) * (arr[:, 2]) * (1.-arr[:, 3]))
baba = ((arr[:, 0]) * (1.-arr[:, 1]) * (arr[:, 2]) * (1.-arr[:, 3]))
top = abba - baba
bot = abba + baba
## get statistic and avoid zero div
sbot = bot.sum()
if sbot != 0:
dst = top.sum() / float(sbot)
else:
dst = 0
return abba.sum(), baba.sum(), dst
@numba.jit(nopython=True)
def _get_boots(arr, nboots):
"""
return array of bootstrap D-stats
"""
## hold results (nboots, [dstat, ])
boots = np.zeros((nboots,))
## iterate to fill boots
for bidx in range(nboots):
## sample with replacement
lidx = np.random.randint(0, arr.shape[0], arr.shape[0])
tarr = arr[lidx]
_, _, dst = _prop_dstat(tarr)
boots[bidx] = dst
## return bootarr
return boots
@numba.jit(nopython=True)
def _get_signif_4(arr, nboots):
"""
returns a list of stats and an array of dstat boots. Stats includes
z-score and two-sided P-value.
"""
abba, baba, dst = _prop_dstat(arr)
boots = _get_boots(arr, nboots)
estimate, stddev = (boots.mean(), boots.std())
zscore = 0.
if stddev:
zscore = np.abs(dst) / stddev
stats = [dst, estimate, stddev, zscore, abba, baba, arr.shape[0]]
return np.array(stats), boots
@numba.jit(nopython=True)
def _get_signif_5(arr, nboots):
"""
returns a list of stats and an array of dstat boots. Stats includes
z-score and two-sided P-value.
"""
statsarr = np.zeros((3, 7), dtype=np.float64)
bootsarr = np.zeros((3, nboots))
idx = 0
for acol in [2, 3, 4]:
rows = np.array([0, 1, acol, 5])
tarr = arr[:, rows, :]
abxa, baxa, dst = _prop_dstat(tarr)
boots = _get_boots(tarr, nboots)
estimate, stddev = (boots.mean(), boots.std())
if stddev:
zscore = np.abs(dst) / stddev
else:
zscore = np.NaN
stats = [dst, estimate, stddev, zscore, abxa, baxa, arr.shape[0]]
statsarr[idx] = stats
bootsarr[idx] = boots
idx += 1
return statsarr, bootsarr
######################################################################
## Simulation functions (requires msprime)
######################################################################
class Sim(object):
def __init__(self, names, sims, nreps, debug):
if not sys.modules.get("msprime"):
raise ImportError(_MSPRIME_IMPORT)
self.names = names
self.sims = sims
self.nreps = nreps
self.debug = debug
def _simulate(self, nreps, admix=None, Ns=500000, gen=20):
"""
Enter a baba.Tree object in which the 'tree' attribute (newick
derived tree) has edge lengths in units of generations. You can
use the 'gen' parameter to multiply branch lengths by a constant.
Parameters:
-----------
nreps: (int)
Number of reps (loci) to simulate under the demographic scenario
tree: (baba.Tree object)
A baba.Tree object initialized by calling baba.Tree(*args).
admix: (list)
A list of admixture events to occur on the tree. Nodes must be
reference by their index number, and events must occur in time
intervals when edges exist. Use the .draw() function of the
baba.Tree object to see node index numbers and coalescent times.
Ns: (float)
Fixed effective population size for all lineages (may allow to vary
in the future).
gen: (int)
A multiplier applied to branch lengths to scale into units of
generations. Example, if all edges on a tree were 1 then you might
enter 50000 to multiply so that edges are 50K generations long.
"""
## node ages
Taus = np.array(list(set(self.verts[:, 1]))) * 1e4 * gen
## The tips samples, ordered alphanumerically
## Population IDs correspond to their indexes in pop config
ntips = len(self.tree)
#names = {name: idx for idx, name in enumerate(sorted(self.tree.get_leaf_names()))}
## rev ladderized leaf name order (left to right on downward facing tree)
names = {name: idx for idx, name in enumerate(self.tree.get_leaf_names()[::-1])}
pop_config = [
ms.PopulationConfiguration(sample_size=2, initial_size=Ns)
for i in range(ntips)
]
## migration matrix all zeros init
migmat = np.zeros((ntips, ntips)).tolist()
## a list for storing demographic events
demog = []
## coalescent times
coals = sorted(list(set(self.verts[:, 1])))[1:]
for ct in range(len(coals)):
## check for admix event before next coalescence
## ...
## print coals[ct], nidxs, time
nidxs = np.where(self.verts[:, 1] == coals[ct])[0]
time = Taus[ct+1]
## add coalescence at each node
for nidx in nidxs:
node = self.tree.search_nodes(name=str(nidx))[0]
## get destionation (lowest child idx number), and other
dest = sorted(node.get_leaves(), key=lambda x: x.idx)[0]
otherchild = [i for i in node.children if not
i.get_leaves_by_name(dest.name)][0]
## get source
if otherchild.is_leaf():
source = otherchild
else:
source = sorted(otherchild.get_leaves(), key=lambda x: x.idx)[0]
## add coal events
event = ms.MassMigration(
time=int(time),
source=names[source.name],
destination=names[dest.name],
proportion=1.0)
#print(int(time), names[source.name], names[dest.name])