Merge pull request #86 from hammerlab/remove-soft-clipped-bases

By default, removing soft-clipped bases.
openvax · Jun 21, 2017 · d2558c8 · d2558c8
2 parents 2a62ce0 + 213baa8
commit d2558c8
Show file tree

Hide file tree

Showing 2 changed files with 14 additions and 2 deletions.
diff --git a/isovar/__init__.py b/isovar/__init__.py
@@ -14,4 +14,4 @@
 
 from __future__ import print_function, division, absolute_import
 
-__version__ = "0.6.1"
+__version__ = "0.7.0"
diff --git a/isovar/locus_reads.py b/isovar/locus_reads.py
@@ -65,7 +65,8 @@ def from_pysam_pileup_element(
             base0_position_after_variant,
             use_secondary_alignments,
             use_duplicate_reads,
-            min_mapping_quality):
+            min_mapping_quality,
+            use_soft_clipped_bases=False):
         """
         Parameters
         ----------
@@ -81,6 +82,8 @@ def from_pysam_pileup_element(
 
         min_mapping_quality : int
 
+        use_soft_clipped_bases : bool. Default false; set to true to keep soft-clipped bases
+
         Returns LocusRead or None
         """
         read = pileup_element.alignment
@@ -190,6 +193,15 @@ def from_pysam_pileup_element(
         if isinstance(sequence, bytes):
             sequence = sequence.decode('ascii')
 
+        if not use_soft_clipped_bases:
+            start = read.query_alignment_start
+            end = read.query_alignment_end
+            sequence = sequence[start:end]
+            reference_positions = reference_positions[start:end]
+            base_qualities = base_qualities[start:end]
+            base0_read_position_before_variant -= start
+            base0_read_position_after_variant -= start
+
         return cls(
             name=name,
             sequence=sequence,