-
Notifications
You must be signed in to change notification settings - Fork 36
/
compute_probs.py
240 lines (199 loc) · 10.7 KB
/
compute_probs.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
#import math
import sys
from basic import *
from util import get_top_genes
#import matplotlib
#if make_png: matplotlib.use('Agg')
#import matplotlib.pyplot as plt
#import numpy as np
import tcr_sampler
from translation import get_translation
from amino_acids import amino_acids
new_probs = pipeline_params['new_probs']
if new_probs:
print 'compute_probs: new_probs'
import tcr_rearrangement_new as tcr_rearrangement ## all_rep_probs
else:
print 'compute_probs: old_probs'
import tcr_rearrangement
with Parser(locals()) as p:
#p.str('args').unspecified_default().multiple().required()
p.str('organism')
p.str('infile').required()
p.str('outfile').required()
#p.int('min_mice').default(2)
#p.float('float_arg') # --float_arg 9.6
#p.flag('plot') # --flag_arg (no argument passed)
p.flag('verbose').shorthand('v') # --flag_arg (no argument passed)
p.flag('allow_stop_codons') # --flag_arg (no argument passed)
p.flag('allow_X') # --flag_arg (no argument passed)
p.flag('clobber').shorthand('c') # --flag_arg (no argument passed)
p.flag('add_masked_seqs') # --flag_arg (no argument passed)
p.flag('filter') # --flag_arg (no argument passed)
p.int('max_cdr3_length_for_filtering').default(30) # --flag_arg (no argument passed)
p.int('min_cdr3_length_for_filtering').default(6) # otherwise compute_distances.py will fail...
p.flag('no_probabilities').described_as('Assign a probability of 1 to all TCRs.')
#p.flag('find_exact_matches') # --flag_arg (no argument passed)
#p.range('range_arg') # --range_arg 1:2
#p.multiword('multi_arg') # --multi_arg hello world
#p.file('file_arg') # --file_arg README.txt
#p.directory('dir_arg') # --dir_arg /tmp/
#p.str('floatlist').cast(lambda x: [float(val) for val in x.split(',')])
#p.multiword('intlist').cast(lambda x: [int(val) for val in x.split()])
assert add_masked_seqs
if exists(outfile):assert clobber
out = open(outfile,'w')
infields = []
outfields = []
for line in open(infile,'rU'):
if not infields:
if line[0] == '#':
infields = line[1:-1].split('\t')
else:
infields = line[:-1].split('\t')
outfields = infields[:]
outfields.extend( ['a_protseq_prob','cdr3a_protseq_prob','va_rep_prob','ja_rep_prob','a_nucseq_prob',
'b_protseq_prob','cdr3b_protseq_prob','vb_rep_prob','jb_rep_prob','b_nucseq_prob' ] )
if add_masked_seqs:
outfields.extend( ['cdr3a_protseq_masked','a_indels','cdr3a_new_nucseq',
'cdr3b_protseq_masked','b_indels','cdr3b_new_nucseq' ] )
out.write('\t'.join( outfields )+'\n' )
continue
assert infields
l = parse_tsv_line( line[:-1], infields )
if filter and 'status' in l and l['status'] != 'OK':continue
theid = line.split("\t")[0]
## ALPHA
va_gene = l['va_gene']
ja_gene = l['ja_gene']
vb_gene = l['vb_gene']
jb_gene = l['jb_gene']
cdr3a_protseq = l['cdr3a']
cdr3a_nucseq = l['cdr3a_nucseq']
cdr3b_protseq = l['cdr3b']
cdr3b_nucseq = l['cdr3b_nucseq']
if filter:
if 'UNK' in va_gene+ja_gene or 'TRa' in va_gene+ja_gene: continue
if 'UNK' in va_gene+ja_gene or 'TRa' in va_gene+ja_gene: continue
if ( len(cdr3a_protseq) > max_cdr3_length_for_filtering or
len(cdr3b_protseq) > max_cdr3_length_for_filtering or
len(cdr3a_protseq) < min_cdr3_length_for_filtering or
len(cdr3b_protseq) < min_cdr3_length_for_filtering ): continue
## check for stop codons
skip_me = False
for a in cdr3a_protseq + cdr3b_protseq:
if a not in amino_acids:
assert a in 'X*'
if ( a == '*' and not allow_stop_codons) or ( a == 'X' and not allow_X ):
Log('{} skipping: badseq: {} {}'.format(theid, cdr3a_protseq,cdr3b_protseq))
skip_me = True
break
if skip_me:
continue
## probs are computed by reps
va_reps = l['va_reps'].split(';')
ja_reps = l['ja_reps'].split(';')
va_countreps = l['va_countreps'].split(';')
ja_countreps = l['ja_countreps'].split(';')
va_cdr3_nucseq = tcr_sampler.get_v_cdr3_nucseq( organism, va_gene )
ja_cdr3_nucseq = tcr_sampler.get_j_cdr3_nucseq( organism, ja_gene )
va_cdr3_protseq,codons = get_translation( va_cdr3_nucseq, '+1' )
ja_cdr3_protseq,codons = get_translation( ja_cdr3_nucseq, '+{}'.format(1+len(ja_cdr3_nucseq)%3))
if no_probabilities or not tcr_rearrangement.probs_data_exist( organism,'A'):
##all probabilities will be set to 1 if this flag is set
aprob_nucseq = 1
aprob_protseq = 1
else:
aprob_nucseq,new_cdr3a_nucseq = tcr_sampler.alpha_cdr3_protseq_probability( theid, organism, va_gene, ja_gene,
cdr3_protseq='',
cdr3_nucseq=cdr3a_nucseq, verbose=verbose,
return_final_cdr3_nucseq=True )
if new_cdr3a_nucseq != cdr3a_nucseq: ## note note note
print 'new_cdr3a_nucseq:',len(new_cdr3a_nucseq),new_cdr3a_nucseq
print 'old_cdr3a_nucseq:',len(cdr3a_nucseq),cdr3a_nucseq
new_cdr3a_protseq = get_translation( new_cdr3a_nucseq, '+1' )[0]
else:
new_cdr3a_protseq = cdr3a_protseq[:]
assert new_cdr3a_protseq == get_translation( cdr3a_nucseq, '+1' )[0]
aprob_protseq = tcr_sampler.alpha_cdr3_protseq_probability( theid, organism, va_gene, ja_gene, new_cdr3a_protseq,
verbose=verbose )
## BETA
vb_reps = l['vb_reps'].split(';')
jb_reps = l['jb_reps'].split(';')
vb_countreps = l['vb_countreps'].split(';')
jb_countreps = l['jb_countreps'].split(';')
vb_cdr3_nucseq = tcr_sampler.get_v_cdr3_nucseq( organism, vb_gene )
jb_cdr3_nucseq = tcr_sampler.get_j_cdr3_nucseq( organism, jb_gene )
vb_cdr3_protseq,codons = get_translation( vb_cdr3_nucseq, '+1' )
jb_cdr3_protseq,codons = get_translation( jb_cdr3_nucseq, '+{}'.format(1+len(jb_cdr3_nucseq)%3))
if no_probabilities or not tcr_rearrangement.probs_data_exist( organism,'B'):
##all probabilities will be set to 1 if this flag is set
bprob_nucseq = 1
bprob_protseq = 1
else:
bprob_nucseq, new_cdr3b_nucseq \
= tcr_sampler.beta_cdr3_protseq_probability( theid, organism, vb_gene, jb_gene, cdr3_protseq='',
verbose=verbose, cdr3_nucseq=cdr3b_nucseq,
allow_early_nucseq_mismatches=True,
return_final_cdr3_nucseq=True )
if new_cdr3b_nucseq != cdr3b_nucseq: ## note note note
new_cdr3b_protseq = get_translation( new_cdr3b_nucseq, '+1' )[0]
else:
new_cdr3b_protseq = cdr3b_protseq[:]
assert new_cdr3b_protseq == get_translation( cdr3b_nucseq, '+1' )[0]
bprob_protseq = tcr_sampler.beta_cdr3_protseq_probability( theid, organism, vb_gene, jb_gene, new_cdr3b_protseq,
verbose=verbose )
vals = dict(l) #line.split('\t') + ['']*(len(outfields)-len(infields))
if add_masked_seqs:
## junction analysis
a_junction_results = tcr_sampler.analyze_junction( organism, va_gene, ja_gene, cdr3a_protseq, cdr3a_nucseq )
b_junction_results = tcr_sampler.analyze_junction( organism, vb_gene, jb_gene, cdr3b_protseq, cdr3b_nucseq )
cdr3a_new_nucseq, cdr3a_protseq_masked, cdr3a_protseq_new_nucleotide_countstring,a_trims,a_inserts \
= a_junction_results
cdr3b_new_nucseq, cdr3b_protseq_masked, cdr3b_protseq_new_nucleotide_countstring,b_trims,b_inserts \
= b_junction_results
# from tcr_sampler.py:
# trims = ( v_trim, d0_trim, d1_trim, j_trim )
# inserts = ( best_d_id, n_vd_insert, n_dj_insert, n_vj_insert )
assert a_trims[1] + a_trims[2] + a_inserts[0] + a_inserts[1] + a_inserts[2] + b_inserts[3] == 0
assert a_inserts[3] == len( cdr3a_new_nucseq )
ita = '+%d-%d'%(sum(a_inserts[1:]),sum(a_trims))
itb = '+%d-%d'%(sum(b_inserts[1:]),sum(b_trims))
vals[ 'cdr3a_protseq_masked'] = cdr3a_protseq_masked
vals[ 'a_indels'] = ita
vals[ 'cdr3a_new_nucseq' ] = cdr3a_new_nucseq
vals[ 'cdr3b_protseq_masked'] = cdr3b_protseq_masked
vals[ 'b_indels'] = itb
vals[ 'cdr3b_new_nucseq' ] = cdr3b_new_nucseq
## there's a little bit of a bias toward guys with more blast hits, ie shorted reads? since we take a max
if no_probabilities or not ( tcr_rearrangement.probs_data_exist(organism,'A') and
tcr_rearrangement.probs_data_exist(organism,'B') ):
##all probabilities will be set to 1 if this flag is set
va_rep_prob = 1
ja_rep_prob = 1
vb_rep_prob = 1
jb_rep_prob = 1
elif new_probs:
va_rep_prob = max( [ tcr_rearrangement.all_countrep_pseudoprobs[organism]['A']['V'][x] for x in va_countreps ] )
ja_rep_prob = max( [ tcr_rearrangement.all_countrep_pseudoprobs[organism]['A']['J'][x] for x in ja_countreps ] )
vb_rep_prob = max( [ tcr_rearrangement.all_countrep_pseudoprobs[organism]['B']['V'][x] for x in vb_countreps ] )
jb_rep_prob = max( [ tcr_rearrangement.all_countrep_pseudoprobs[organism]['B']['J'][x] for x in jb_countreps ] )
else:
va_rep_prob = max( [ tcr_rearrangement.all_rep_probs[organism][x] for x in va_reps ] )
ja_rep_prob = max( [ tcr_rearrangement.all_rep_probs[organism][x] for x in ja_reps ] )
vb_rep_prob = max( [ tcr_rearrangement.all_rep_probs[organism][x] for x in vb_reps ] )
jb_rep_prob = max( [ tcr_rearrangement.all_rep_probs[organism][x] for x in jb_reps ] )
vals['a_protseq_prob' ] = aprob_protseq * va_rep_prob * ja_rep_prob
vals['cdr3a_protseq_prob'] = aprob_protseq
vals['va_rep_prob' ] = va_rep_prob
vals['ja_rep_prob' ] = ja_rep_prob
vals['a_nucseq_prob' ] = aprob_nucseq * va_rep_prob * ja_rep_prob
vals['b_protseq_prob' ] = bprob_protseq * vb_rep_prob * jb_rep_prob
vals['cdr3b_protseq_prob'] = bprob_protseq
vals['vb_rep_prob' ] = vb_rep_prob
vals['jb_rep_prob' ] = jb_rep_prob
vals['b_nucseq_prob' ] = bprob_nucseq * vb_rep_prob * jb_rep_prob
assert len(vals.keys()) == len(outfields)
out.write( make_tsv_line( vals, outfields, '-' )+'\n' )
out.flush()
out.close()