Merge pull request #36 from stjude/fix

Fix

rnaindel/analysis/alignment_feature_calculator.py

@@ -206,7 +206,10 @@ def make_indel_alignment(variant, bam, downsample_threshold=1500):
 
  try:
  valn = VariantAlignment(
- variant, bam, downsample_threshold=downsample_threshold, exclude_duplicates=True
+ variant,
+ bam,
+ downsample_threshold=downsample_threshold,
+ exclude_duplicates=True,
  )
  except:
  return None

rnaindel/occurrence/counter.py

@@ -2,6 +2,7 @@
 
 import os
 import sys
+import gzip
 import pysam
 import argparse
 from functools import partial
@@ -61,7 +62,15 @@ def validate_file_lst(filename):
 
 
 def is_rnaindel_output_vcf(vcf):
- headers = [line for line in open(vcf) if line.startswith("##source=RNAIndel")]
+ if vcf.endswith(".gz"):
+ headers = [
+ line
+ for line in gzip.open(vcf, "rt")
+ if line.startswith("##source=RNAIndel")
+ ]
+ else:
+ headers = [line for line in open(vcf) if line.startswith("##source=RNAIndel")]
+
  if headers:
  return True
  else:
@@ -93,7 +102,15 @@ def make_indel_from_vcf_line(line, genome):
 
 
 def collect_somatic_predictions_from_vcf(vcf, genome):
- somatic_lines = [line for line in open(vcf) if "predicted_class=somatic" in line]
+ if vcf.endswith(".gz"):
+ somatic_lines = [
+ line for line in gzip.open(vcf, "rt") if "predicted_class=somatic" in line
+ ]
+ else:
+ somatic_lines = [
+ line for line in open(vcf) if "predicted_class=somatic" in line
+ ]
+
  parsed_somatic_lines = [
  make_indel_from_vcf_line(line, genome)
  for line in somatic_lines
@@ -102,10 +119,19 @@ def collect_somatic_predictions_from_vcf(vcf, genome):
  return to_flat_lst(parsed_somatic_lines)
 
 
+def rreplace(s, old, new):
+ li = s.rsplit(old, 1)
+ return new.join(li)
+
+
 def annotate_vcf_with_recurrence(vcf, genome, occurrence_dict, outdir):
  new_vcf = edit_header(vcf)
 
- fi = open(vcf)
+ if vcf.endswith(".gz"):
+ fi = gzip.open(vcf, "rt")
+ else:
+ fi = open(vcf)
+
  for line in fi:
  if line.startswith("#"):
  pass
@@ -120,18 +146,33 @@ def annotate_vcf_with_recurrence(vcf, genome, occurrence_dict, outdir):
  fi.close()
 
  if outdir:
- filename = os.path.basename(vcf).replace(".vcf", ".occurrence.vcf")
- fo = open(os.path.join(outdir, filename), "w")
+ filename = os.path.join(outdir, os.path.basename(vcf))
  else:
- fo = open(vcf, "w")
+ filename = vcf
+
+ filename = rreplace(filename, ".vcf", ".occurrence.vcf")
+ filename = rreplace(filename, ".gz", "")
+
+ fo = open(filename, "w")
+
  fo.write("".join(new_vcf))
  fo.close()
 
+ pysam.tabix_index(filename, preset="vcf", force=True, keep_original=False)
+
 
 def edit_header(vcf):
- header_lines = [line for line in open(vcf) if line.startswith("##")]
+ if vcf.endswith(".gz"):
+ header_lines = [line for line in gzip.open(vcf, "rt") if line.startswith("##")]
+ else:
+ header_lines = [line for line in open(vcf) if line.startswith("##")]
+
  new_header_line = '##INFO=<ID=occurrence,Number=1,Type=Integer,Description="Occurrence count in the input VCF files. Only counted for indels predicted as somatic">\n'
- bottom = [line for line in open(vcf) if line.startswith("#CHROM")]
+
+ if vcf.endswith(".gz"):
+ bottom = [line for line in gzip.open(vcf, "rt") if line.startswith("#CHROM")]
+ else:
+ bottom = [line for line in open(vcf) if line.startswith("#CHROM")]
 
  if new_header_line not in header_lines:
  return header_lines + [new_header_line] + bottom

rnaindel/version.py

@@ -1 +1,2 @@
-__version__ = "3.2.2"
+__version__ = "3.2.3"
+