Merge pull request #1198 from griffithlab/pvacsplice_include_nonhuman…

…EnsemblTranscript include non-human transcripts to pvacsplice
griffithlab · Feb 19, 2025 · 7dc4386 · 7dc4386
2 parents bfbf2b5 + cbb920a
commit 7dc4386
Show file tree

Hide file tree

Showing 3 changed files with 3 additions and 2 deletions.
diff --git a/pvactools/lib/combine_inputs.py b/pvactools/lib/combine_inputs.py
@@ -15,7 +15,7 @@ def add_junction_coordinates_to_variants(self):
 
         # remove version number in annotated to compare with filtered junctions file
         var_df[['transcript_id', 'transcript_version']] = var_df['transcript_name'].str.split('.', expand=True)
-        var_df = var_df.loc[var_df['transcript_id'].str.startswith('ENST') == True]
+        var_df = var_df.loc[var_df['transcript_id'].fillna('').str.contains(r'^ENS.*T', regex=True)]
         var_df['transcript_version'] = var_df['transcript_version'].astype('int64')
 
         # create new cols

diff --git a/pvactools/lib/junction_to_fasta.py b/pvactools/lib/junction_to_fasta.py
@@ -161,7 +161,7 @@ def get_aa_sequence(self, dataframe):
         final_seq = ''
         for x in coordinates:
             start = int(x.split(',')[0]); end = int(x.split(',')[1])
-            seq = self.personal_fasta.get_seq(self.chrom, start, end)
+            seq = self.personal_fasta.get_seq(str(self.chrom), start, end)
             final_seq += str(seq)
         # using Seq from Bio.Seq to translate str_seq
         # positive strand

diff --git a/pvactools/lib/splice_pipeline.py b/pvactools/lib/splice_pipeline.py
@@ -1,3 +1,4 @@
+import sys
 import shutil
 import os
 import pandas as pd