Merge pull request #336 from Hoeze/aparent_v3

fix missing reverse-complement in reference sequence fetching for APARENT
kipoi · Jul 4, 2022 · ae8cf12 · ae8cf12
2 parents 77fa3d8 + 61f1f93
commit ae8cf12
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 3 deletions.
diff --git a/APARENT/site_probabilities/dataloader.py b/APARENT/site_probabilities/dataloader.py
@@ -21,7 +21,9 @@ def __init__(
         self.reference_sequence = reference_sequence
         self.interval_attrs = interval_attrs
 
-        self.variant_seq_extractor = VariantSeqExtractor(reference_sequence=reference_sequence)
+        if not self.reference_sequence.use_strand:
+            raise ValueError(
+                "Reference sequence fetcher does not use strand but this is needed to obtain correct sequences!")
 
         self.one_hot = OneHot()
 
@@ -134,5 +136,5 @@ def __init__(
 
         super().__init__(
             regions_of_interest=roi,
-            reference_sequence=FastaStringExtractor(fasta_file),
+            reference_sequence=FastaStringExtractor(fasta_file, use_strand=True),
         )
diff --git a/APARENT/veff/dataloader.py b/APARENT/veff/dataloader.py
@@ -23,6 +23,9 @@ def __init__(
         self.variants = variants
         self.interval_attrs = interval_attrs
 
+        if not self.reference_sequence.use_strand:
+            raise ValueError(
+                "Reference sequence fetcher does not use strand but this is needed to obtain correct sequences!")
         self.variant_seq_extractor = VariantSeqExtractor(reference_sequence=reference_sequence)
 
         self.matcher = SingleVariantMatcher(
@@ -155,6 +158,6 @@ def __init__(
         from kipoiseq.extractors import MultiSampleVCF
         super().__init__(
             regions_of_interest=roi,
-            reference_sequence=FastaStringExtractor(fasta_file),
+            reference_sequence=FastaStringExtractor(fasta_file, use_strand=True),
             variants=MultiSampleVCF(vcf_file, lazy=vcf_lazy)
         )