Skip to content

Commit 4152bac

Browse files
drosofffdrosofff
authored
Add svtyper-python3 0.7.1 (bioconda#60004)
* Create meta.yml recipe for svtyper-python3 * change recipe with recommended patch strategy * Create python3-compatibility.patch
1 parent 9a5f128 commit 4152bac

2 files changed

Lines changed: 275 additions & 0 deletions

File tree

‎recipes/svtyper-python3/meta.yml‎

Lines changed: 49 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,49 @@
1+
{% set name = "svtyper-python3" %}
2+
{% set version = "0.7.1" %}
3+
4+
package:
5+
name: {{ name|lower }}
6+
version: {{ version }}
7+
8+
source:
9+
# Using the official upstream release
10+
url: https://github.com/hall-lab/svtyper/archive/v{{ version }}.tar.gz
11+
sha256: 0abf90172eecad49511a14c9c383314e973cac7a7cc086ed88b7757f00cce0b4
12+
# Applying a patch for Python 3 compatibility from brentp's PR
13+
# https://github.com/hall-lab/svtyper/pull/110
14+
patches:
15+
- python3-compatibility.patch
16+
17+
build:
18+
number: 0
19+
noarch: python
20+
script: "{{ PYTHON }} -m pip install . --no-deps --ignore-installed -vv"
21+
entry_points:
22+
- svtyper = svtyper.cli:main
23+
24+
requirements:
25+
host:
26+
- python >=3.6
27+
- pip
28+
- setuptools
29+
run:
30+
- python >=3.6
31+
- pysam >=0.15.0
32+
- numpy
33+
- cytoolz
34+
35+
test:
36+
imports:
37+
- svtyper
38+
commands:
39+
- svtyper --help
40+
41+
about:
42+
home: https://github.com/hall-lab/svtyper
43+
summary: 'Bayesian genotyper for structural variants'
44+
license: MIT
45+
license_file: LICENSE
46+
47+
extra:
48+
recipe-maintainers:
49+
- drosofff

‎recipes/svtyper-python3/python3-compatibility.patch‎

Lines changed: 226 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,226 @@
1+
diff --git a/svtyper/classic.py b/svtyper/classic.py
2+
index 76dc97c..97561f9 100755
3+
--- a/svtyper/classic.py
4+
+++ b/svtyper/classic.py
5+
@@ -13,6 +13,10 @@ from svtyper.parsers import Vcf, Variant, Sample, confidence_interval
6+
from svtyper.utils import *
7+
from svtyper.statistics import bayes_gt
8+
9+
+try:
10+
+ xrange
11+
+except NameError:
12+
+ xrange = range
13+
# --------------------------------------
14+
# define functions
15+
16+
@@ -126,7 +130,7 @@ def sv_genotype(bam_string,
17+
bam_list.append(pysam.AlignmentFile(b, mode='rb'))
18+
elif b.endswith('.cram'):
19+
bam_list.append(pysam.AlignmentFile(b,
20+
- mode='rc',reference_filename=ref_fasta,format_options=["required_fields=7167"]))
21+
+ mode='rc',reference_filename=ref_fasta,format_options=[b"required_fields=7167"]))
22+
else:
23+
sys.stderr.write('Error: %s is not a valid alignment file (*.bam or *.cram)\n' % b)
24+
exit(1)
25+
@@ -413,12 +417,12 @@ def sv_genotype(bam_string,
26+
out_bam_written_reads = write_alignment(read, out_bam, out_bam_written_reads)
27+
28+
if debug:
29+
- print '--------------------------'
30+
- print 'ref_span:', ref_span
31+
- print 'alt_span:', alt_span
32+
- print 'ref_seq:', ref_seq
33+
- print 'alt_seq:', alt_seq
34+
- print 'alt_clip:', alt_clip
35+
+ print('--------------------------')
36+
+ print('ref_span:', ref_span)
37+
+ print('alt_span:', alt_span)
38+
+ print('ref_seq:', ref_seq)
39+
+ print('alt_seq:', alt_seq)
40+
+ print('alt_clip:', alt_clip)
41+
42+
# in the absence of evidence for a particular type, ignore the reference
43+
# support for that type as well
44+
@@ -443,12 +447,12 @@ def sv_genotype(bam_string,
45+
QR = int(split_weight * ref_seq) + int(disc_weight * ref_span)
46+
QA = int(split_weight * alt_splitters) + int(disc_weight * alt_span)
47+
gt_lplist = bayes_gt(QR, QA, is_dup)
48+
- best, second_best = sorted([ (i, e) for i, e in enumerate(gt_lplist) ], key=lambda(x): x[1], reverse=True)[0:2]
49+
+ best, second_best = sorted([ (i, e) for i, e in enumerate(gt_lplist) ], key=lambda x: x[1], reverse=True)[0:2]
50+
gt_idx = best[0]
51+
52+
# print log probabilities of homref, het, homalt
53+
if debug:
54+
- print gt_lplist
55+
+ print(gt_lplist)
56+
57+
# set the overall variant QUAL score and sample specific fields
58+
var.genotype(sample.name).set_format('GL', ','.join(['%.0f' % x for x in gt_lplist]))
59+
@@ -573,7 +577,7 @@ def main():
60+
def cli():
61+
try:
62+
sys.exit(main())
63+
- except IOError, e:
64+
+ except IOError as e:
65+
if e.errno != 32: # ignore SIGPIPE
66+
raise
67+
68+
diff --git a/svtyper/parsers.py b/svtyper/parsers.py
69+
index 1357b8c..6f116ec 100644
70+
--- a/svtyper/parsers.py
71+
+++ b/svtyper/parsers.py
72+
@@ -3,15 +3,16 @@ from __future__ import print_function
73+
import time, re, json, sys
74+
from collections import Counter
75+
76+
-from svtyper.statistics import mean, stdev, median, upper_mad
77+
+from svtyper.statistics import mean, stdev, median, upper_mad, xrange
78+
+
79+
80+
# ==================================================
81+
# VCF parsing tools
82+
# ==================================================
83+
def confidence_interval(var, tag, alt_tag, max_ci_dist):
84+
- ci = map(int, var.info[tag].split(','))
85+
+ ci = list(map(int, var.info[tag].split(',')))
86+
if ci[1] - ci[0] > max_ci_dist:
87+
- return map(int, var.info[alt_tag].split(','))
88+
+ return list(map(int, var.info[alt_tag].split(',')))
89+
return ci
90+
91+
92+
@@ -480,7 +481,7 @@ class Library(object):
93+
for r in bam.header['RG']:
94+
try:
95+
in_lib = r['LB'] == lib_name
96+
- except KeyError, e:
97+
+ except KeyError as e:
98+
in_lib = lib_name == ''
99+
100+
if in_lib:
101+
@@ -520,8 +521,10 @@ class Library(object):
102+
for read in self.bam.fetch():
103+
if read.get_tag('RG') not in self.readgroups:
104+
continue
105+
- if read.infer_query_length() > max_rl:
106+
- max_rl = read.infer_query_length()
107+
+ rl = read.infer_query_length()
108+
+ if rl is None: continue
109+
+ if rl > max_rl:
110+
+ max_rl = rl
111+
if counter == num_samp:
112+
break
113+
counter += 1
114+
@@ -669,7 +672,7 @@ class Sample(object):
115+
for r in bam.header['RG']:
116+
try:
117+
lib_name=r['LB']
118+
- except KeyError, e:
119+
+ except KeyError as e:
120+
lib_name=''
121+
122+
# add the new library
123+
@@ -877,7 +880,7 @@ class SamFragment(object):
124+
try:
125+
p = float(self.lib.dens[ospan_length]) * conc_prior / (conc_prior * self.lib.dens[ospan_length] + disc_prior * (self.lib.dens[ospan_length - var_length]))
126+
except ZeroDivisionError:
127+
- p = None
128+
+ return False
129+
130+
return p > 0.5
131+
132+
diff --git a/svtyper/singlesample.py b/svtyper/singlesample.py
133+
index 4acb845..d3b16b0 100644
134+
--- a/svtyper/singlesample.py
135+
+++ b/svtyper/singlesample.py
136+
@@ -47,17 +47,33 @@ description: Compute genotype of structural variants based on breakpoint depth o
137+
return args
138+
139+
def ensure_valid_alignment_file(afile):
140+
- if not (afile.endswith('.bam') or afile.endswith('.cram')):
141+
- die('Error: %s is not a valid alignment file (*.bam or *.cram)\n' % afile)
142+
+ print(afile.__class__.__name__, file=sys.stderr)
143+
+ if not isinstance(afile, str):
144+
+ afile = afile.decode()
145+
+
146+
+ try:
147+
+ if not (afile.endswith('.bam') or afile.endswith('.cram')):
148+
+ die('Error: %s is not a valid alignment file (*.bam or *.cram)\n' % afile)
149+
+ except TypeError:
150+
+ if not (afile.endswith(b'.bam') or afile.endswith(b'.cram')):
151+
+ die('Error: %s is not a valid alignment file (*.bam or *.cram)\n' % afile)
152+
153+
def open_alignment_file(afile, reference_fasta):
154+
fd = None
155+
- if afile.endswith('.bam'):
156+
- fd = pysam.AlignmentFile(afile, mode='rb')
157+
- elif afile.endswith('.cram'):
158+
- fd = pysam.AlignmentFile(afile, mode='rc', reference_filename=reference_fasta)
159+
- else:
160+
- die('Error: %s is not a valid alignment file (*.bam or *.cram)\n' % afile)
161+
+ try:
162+
+ if afile.endswith('.bam'):
163+
+ fd = pysam.AlignmentFile(afile, mode='rb')
164+
+ elif afile.endswith('.cram'):
165+
+ fd = pysam.AlignmentFile(afile, mode='rc', reference_filename=reference_fasta)
166+
+ else:
167+
+ die('Error: %s is not a valid alignment file (*.bam or *.cram)\n' % afile)
168+
+ except TypeError:
169+
+ if afile.endswith(b'.bam'):
170+
+ fd = pysam.AlignmentFile(afile, mode='rb')
171+
+ elif afile.endswith(b'.cram'):
172+
+ fd = pysam.AlignmentFile(afile, mode='rc', reference_filename=reference_fasta)
173+
+ else:
174+
+ die('Error: %s is not a valid alignment file (*.bam or *.cram)\n' % afile)
175+
176+
return fd
177+
178+
@@ -417,7 +433,7 @@ def bayesian_genotype(breakpoint, counts, split_weight, disc_weight, debug):
179+
180+
# the actual bayesian calculation and decision
181+
gt_lplist = bayes_gt(QR, QA, is_dup)
182+
- best, second_best = sorted([ (i, e) for i, e in enumerate(gt_lplist) ], key=lambda(x): x[1], reverse=True)[0:2]
183+
+ best, second_best = sorted([ (i, e) for i, e in enumerate(gt_lplist) ], key=lambda x: x[1], reverse=True)[0:2]
184+
gt_idx = best[0]
185+
186+
# print log probabilities of homref, het, homalt
187+
@@ -846,7 +862,7 @@ def main():
188+
def cli():
189+
try:
190+
sys.exit(main())
191+
- except IOError, e:
192+
+ except IOError as e:
193+
if e.errno != 32: # ignore SIGPIPE
194+
raise
195+
196+
diff --git a/svtyper/statistics.py b/svtyper/statistics.py
197+
index 51cc3cd..1e005a2 100644
198+
--- a/svtyper/statistics.py
199+
+++ b/svtyper/statistics.py
200+
@@ -1,6 +1,11 @@
201+
import math
202+
from collections import Counter
203+
204+
+try:
205+
+ xrange
206+
+except NameError:
207+
+ xrange = range
208+
+
209+
# ==================================================
210+
# Statistical tools
211+
# ==================================================
212+
diff --git a/svtyper/utils.py b/svtyper/utils.py
213+
index e80793e..f3fac57 100644
214+
--- a/svtyper/utils.py
215+
+++ b/svtyper/utils.py
216+
@@ -47,6 +47,10 @@ def write_sample_json(sample_list, lib_info_file):
217+
lib_info[sample.name] = s
218+
219+
# write the json file
220+
+ for k, sample in lib_info.items():
221+
+ if isinstance(sample["bam"], bytes):
222+
+ lib_info[k]["bam"] = sample["bam"].decode()
223+
+
224+
json.dump(lib_info, lib_info_file, indent=4)
225+
lib_info_file.close()
226+

0 commit comments

Comments
 (0)