* Fixed bug where kallisto quantification would sometimes fail to sum…

… transcripts to genes properly.
GuyTeichman · Oct 3, 2024 · 770db8f · 770db8f
1 parent 8f5da01
commit 770db8f
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 3 deletions.
diff --git a/HISTORY.rst b/HISTORY.rst
@@ -2,6 +2,12 @@
 History
 =======
 
+4.?.? (2024-??-??)
+-------------------
+
+Fixed
+******
+* Fixed bug where kallisto quantification would sometimes fail to sum transcripts to genes properly.
 
 4.1.0 (2024-09-16)
 -------------------

diff --git a/rnalysis/fastq.py b/rnalysis/fastq.py
@@ -2016,16 +2016,15 @@ def _sum_transcripts_to_genes(tpm: pl.DataFrame, counts: pl.DataFrame, gtf_path:
                             pl.exclude(counts.columns[0]).sum().truediv(10 ** 6)).collect()
                         tpm_cpy = tpm.lazy().join(transcript2gene, left_on=tpm.columns[0], right_on='Transcript ID',
                                                   how='left')
-                        tpm_by_gene = tpm_cpy.drop(cs.first()).group_by('Gene ID').sum()
+                        tpm_by_gene = tpm_cpy.drop(cs.first()).drop_nulls().group_by('Gene ID').sum()
                         count_per_gene = tpm_by_gene.with_columns(
                             [(pl.col(col) * library_sizes[col][0]).alias(col) for col in tpm.columns[1:]]).collect()
                     elif summation_method == 'raw':
                         count_cpy = counts.lazy().join(transcript2gene, left_on=tpm.columns[0],
                                                        right_on='Transcript ID', how='left')
-                        count_per_gene = count_cpy.drop(cs.first()).group_by('Gene ID').sum().collect()
+                        count_per_gene = count_cpy.drop(cs.first()).drop_nulls().group_by('Gene ID').sum().collect()
                     else:
                         raise ValueError(f"Invalid value for 'summation_method': '{summation_method}'.")
-
                     if len(count_per_gene) == 0:
                         continue
                     pbar.update(8)