feat: Release v17.1.0 (#1593)

#### Changed

- changed strand bias filter for WGS TO SNV quality filter to only require representation from both strands in alt allele
- lowered maximum raredisease loqusdb SNV observation frequency from 0.01 to 0.007, and verified no common cancer predisposition variant exists above that threshold

Author: mathiasbio

BALSAMIC/constants/cluster_analysis.json

@@ -292,11 +292,11 @@
 		"n": 4
 	},
 	"cadd_annotate_somaticINDEL_research": {
-		"time": "18:00:00",
-		"n": 4
+		"time": "38:00:00",
+		"n": 12
 	},
 	"vep_annotate_somaticSNV_research": {
-		"time": "18:00:00",
+		"time": "28:00:00",
 		"n": 12
 	},
 	"vcfanno_annotate_somaticSNV_clinical": {

BALSAMIC/constants/variant_filters.py

@@ -286,7 +286,7 @@ class WgsSNVFilters(BaseSNVFilters):
         VCFFilter(tag_value=0.001, filter_name="balsamic_high_pop_freq", field="INFO"),
     ]
     clinical = research + [
-        VCFFilter(tag_value=0.01, filter_name="Frq", field="INFO"),
+        VCFFilter(tag_value=0.007, filter_name="Frq", field="INFO"),
         VCFFilter(tag_value=0.1, filter_name="ArtefactFrq", field="INFO"),
     ]
     quality = [

BALSAMIC/snakemake_rules/variant_calling/snv_quality_filter.rule

@@ -36,7 +36,7 @@ bcftools view --threads {threads} --regions-file {input.wgs_calling_file}.bed {i
 | bcftools filter --threads {threads} --include 'FORMAT/AD[0:1] > {params.AD[0]}' --soft-filter '{params.AD[1]}' --mode '+' \
 | bcftools filter --threads {threads} --include 'FORMAT/AF > {params.AF_min[0]}' --soft-filter '{params.AF_min[1]}' --mode '+' \
 | bcftools filter --threads {threads} --include 'SUM(FORMAT/QSS)/SUM(FORMAT/AD) >= {params.qss[0]}' --soft-filter '{params.qss[1]}' --mode '+' \
-| bcftools filter --threads {threads} --include 'FORMAT/ALT_F1R2 > {params.strand_reads[0]} && (FORMAT/ALT_F1R2 > 0  && FORMAT/ALT_F2R1 > {params.strand_reads[0]} &&  FORMAT/REF_F1R2 > {params.strand_reads[0]} && FORMAT/REF_F2R1 > {params.strand_reads[0]})' --soft-filter '{params.strand_reads[1]}' --mode '+' \
+| bcftools filter --threads {threads} --include 'FORMAT/ALT_F1R2 > {params.strand_reads[0]} && FORMAT/ALT_F2R1 > {params.strand_reads[0]}' --soft-filter '{params.strand_reads[1]}' --mode '+' \
 | bcftools filter --threads {threads} --include "INFO/SOR < {params.sor[0]}" --soft-filter '{params.sor[1]}' --mode '+' \
 | bcftools view --exclude '{params.hard_filters}' -o {output.vcf_filtered};
 

CHANGELOG.rst

@@ -1,3 +1,21 @@
+[17.1.0]
+--------
+
+Added:
+^^^^^^
+
+Changed:
+^^^^^^^^
+* removing strand balance requirement of ref-allele in WGS TO strand bias filter https://github.com/Clinical-Genomics/BALSAMIC/pull/1594
+* lowered rare-disease germline loqusdb observation frequency for WGS SNVs to 0.007 from 0.01 https://github.com/Clinical-Genomics/BALSAMIC/pull/1594
+
+Removed:
+^^^^^^^^
+
+Fixed:
+^^^^^^
+
+
 [17.0.1]
 --------
 

docs/balsamic_filters.rst

@@ -647,12 +647,11 @@ The somatic variants in TNscope raw VCF file (`SNV.somatic.<CASE_ID>.tnscope.all
 
     SUM(QSS)/SUM(AD) >= 20
 
-*Read Counts*: Count of reads in a given (F1R2, F2R1) pair orientation supporting the alternate allele and reference alleles
+*Read Counts*: Count of reads in a given (F1R2, F2R1) pair orientation supporting the alternate allele
 
 ::
 
     ALT_F1R2 > 0, ALT_F2R1 > 0
-    REF_F1R2 > 0, REF_F2R1 > 0
 
 *SOR*: Symmetric Odds Ratio of 2x2 contingency table to detect strand bias
 
@@ -682,7 +681,7 @@ This section contains population database frequency filters.
 
 ::
 
-    Frq <= 0.01  (or) Frq == "."
+    Frq <= 0.007  (or) Frq == "."
 
 The variants scored as `PASS` or `triallelic_sites` are included in the final vcf file (`SNV.somatic.<CASE_ID>.tnscope.<research/clinical>.filtered.pass.vcf.gz`).