feat: remove max af 1 from all workflows (#1338)

Removed: - Removed bcftools command for setting soft-filter with maximum allele-frequency, which eventually leads to a hard removal of these variants.
Clinical-Genomics · Jan 4, 2024 · 7397351 · 7397351
1 parent f318b1e
commit 7397351
Show file tree

Hide file tree

Showing 4 changed files with 1 addition and 12 deletions.
diff --git a/BALSAMIC/constants/variant_filters.py b/BALSAMIC/constants/variant_filters.py
@@ -19,7 +19,6 @@
         "field": "INFO",
     },
     "MQ": {"tag_value": 40, "filter_name": "balsamic_low_mq", "field": "INFO"},
-    "AF_max": {"tag_value": 1, "filter_name": "balsamic_af_one", "field": "INFO"},
     "AF_min": {"tag_value": 0.007, "filter_name": "balsamic_low_af", "field": "INFO"},
     "pop_freq": {
         "tag_value": 0.005,
@@ -50,7 +49,6 @@
         "filter_name": "balsamic_low_tumor_dp",
         "field": "FORMAT",
     },
-    "AF_max": {"tag_value": 1, "filter_name": "balsamic_af_one", "field": "FORMAT"},
     "AF_min": {"tag_value": 0.05, "filter_name": "balsamic_low_af", "field": "FORMAT"},
     "pop_freq": {
         "tag_value": 0.001,

diff --git a/BALSAMIC/models/params.py b/BALSAMIC/models/params.py
@@ -194,7 +194,6 @@ class VarCallerFilter(BaseModel):
     Attributes:
         AD: VCFAttributes (required); minimum allelic depth
         AF_min: VCFAttributes (optional); minimum allelic fraction
-        AF_max: VCFAttributes (optional); maximum allelic fraction
         MQ: VCFAttributes (optional); minimum mapping quality
         DP: VCFAttributes (optional); minimum read depth
         pop_freq: VCFAttributes (optional); maximum gnomad allele frequency
@@ -214,7 +213,6 @@ class VarCallerFilter(BaseModel):
 
     AD: Optional[VCFAttributes] = None
     AF_min: Optional[VCFAttributes] = None
-    AF_max: Optional[VCFAttributes] = None
     MQ: Optional[VCFAttributes] = None
     DP: Optional[VCFAttributes] = None
     pop_freq: Optional[VCFAttributes] = None

diff --git a/BALSAMIC/snakemake_rules/variant_calling/sentieon_quality_filter.rule b/BALSAMIC/snakemake_rules/variant_calling/sentieon_quality_filter.rule
@@ -16,7 +16,6 @@ if config["analysis"]["sequencing_type"] == 'wgs' and config["analysis"]["analys
         DP = [SENTIEON_CALLER.DP.tag_value, SENTIEON_CALLER.DP.filter_name],
         AD = [SENTIEON_CALLER.AD.tag_value, SENTIEON_CALLER.AD.filter_name],
         AF_min = [SENTIEON_CALLER.AF_min.tag_value, SENTIEON_CALLER.AF_min.filter_name],
-        AF_max = [SENTIEON_CALLER.AF_max.tag_value, SENTIEON_CALLER.AF_max.filter_name],
         strand_reads = [SENTIEON_CALLER.strand_reads.tag_value, SENTIEON_CALLER.strand_reads.filter_name],
         qss = [SENTIEON_CALLER.qss.tag_value, SENTIEON_CALLER.qss.filter_name],
         sor = [SENTIEON_CALLER.sor.tag_value, SENTIEON_CALLER.sor.filter_name],
@@ -36,7 +35,6 @@ bcftools view -f PASS,triallelic_site --threads {threads} --regions-file {input.
 | bcftools filter --threads {threads} --include 'SUM(FORMAT/AD[0:0]+FORMAT/AD[0:1]) >= {params.DP[0]}' --soft-filter '{params.DP[1]}' --mode '+' \
 | 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 'FORMAT/AF < {params.AF_max[0]}' --soft-filter '{params.AF_max[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 "INFO/SOR < {params.sor[0]}" --soft-filter '{params.sor[1]}' --mode '+' \
@@ -60,7 +58,6 @@ elif config["analysis"]["sequencing_type"] == 'wgs' and config["analysis"]["anal
         AD = [SENTIEON_CALLER.AD.tag_value, SENTIEON_CALLER.AD.filter_name],
         DP = [SENTIEON_CALLER.DP.tag_value, SENTIEON_CALLER.DP.filter_name],
         AF_min = [SENTIEON_CALLER.AF_min.tag_value, SENTIEON_CALLER.AF_min.filter_name],
-        AF_max = [SENTIEON_CALLER.AF_max.tag_value, SENTIEON_CALLER.AF_max.filter_name],
         case_name = config["analysis"]["case_id"],
       threads:
         get_threads(cluster_config, 'bcftools_quality_filter_tnscope_tumor_normal')
@@ -72,7 +69,6 @@ bcftools view -f PASS,triallelic_site {input.vcf_snv} \
 | bcftools filter --threads {threads} --include 'SUM(FORMAT/AD[0:0]+FORMAT/AD[0:1]) >= {params.DP[0]} || SUM(FORMAT/AD[1:0]+FORMAT/AD[1:1]) >= {params.DP[0]}' --soft-filter '{params.DP[1]}' --mode '+' \
 | 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[0] >= {params.AF_min[0]}' --soft-filter '{params.AF_min[1]}' --mode '+' \
-| bcftools filter --threads {threads} --include 'FORMAT/AF[0] <  {params.AF_max[0]}' --soft-filter '{params.AF_max[1]}' --mode '+' \
 | bcftools view -f PASS,triallelic_site -O z -o {output.vcf_snv_research}; 
 
 tabix -p vcf -f {output.vcf_snv_research};
@@ -94,7 +90,6 @@ if config["analysis"]["sequencing_type"] == 'targeted' and config["analysis"]["a
       AD=[VARDICT.AD.tag_value, VARDICT.AD.filter_name],
       DP=[VARDICT.DP.tag_value, VARDICT.DP.filter_name],
       AF_min=[VARDICT.AF_min.tag_value, VARDICT.AF_min.filter_name],
-      AF_max=[VARDICT.AF_max.tag_value, VARDICT.AF_max.filter_name],
       case_name = config["analysis"]["case_id"],
     threads:
       get_threads(cluster_config,'bcftools_quality_filter_vardict_tumor_only')
@@ -107,7 +102,6 @@ bcftools filter --include 'INFO/MQ >= {params.MQ[0]}' --soft-filter '{params.MQ[
 bcftools filter --include 'INFO/DP >= {params.DP[0]}' --soft-filter '{params.DP[1]}' --mode '+' | \
 bcftools filter --include 'INFO/VD >= {params.AD[0]}' --soft-filter '{params.AD[1]}' --mode '+' | \
 bcftools filter --include 'INFO/AF >= {params.AF_min[0]}' --soft-filter '{params.AF_min[1]}' --mode '+' | \
-bcftools filter --include 'INFO/AF <  {params.AF_max[0]}' --soft-filter '{params.AF_max[1]}' --mode '+' | \
 bcftools view -f PASS -o {output.vcf_filtered} -O z;
 
 tabix -p vcf -f {output.vcf_filtered};
@@ -154,7 +148,6 @@ elif config["analysis"]["sequencing_type"] == 'targeted' and config["analysis"][
       AD=[VARDICT.AD.tag_value, VARDICT.AD.filter_name],
       DP=[VARDICT.DP.tag_value, VARDICT.DP.filter_name],
       AF_min=[VARDICT.AF_min.tag_value, VARDICT.AF_min.filter_name],
-      AF_max=[VARDICT.AF_max.tag_value, VARDICT.AF_max.filter_name],
       possible_germline="balsamic_possible_germline",
       case_name = config["analysis"]["case_id"],
     threads:
@@ -168,7 +161,6 @@ bcftools filter --include 'SMPL_MIN(FMT/MQ) >= {params.MQ[0]}' --soft-filter '{p
 bcftools filter --include 'INFO/DP >= {params.DP[0]}' --soft-filter '{params.DP[1]}' --mode '+' | \
 bcftools filter --include 'INFO/VD >= {params.AD[0]}' --soft-filter '{params.AD[1]}' --mode '+' | \
 bcftools filter --include 'INFO/AF >= {params.AF_min[0]}' --soft-filter '{params.AF_min[1]}' --mode '+' | \
-bcftools filter --include 'INFO/AF <  {params.AF_max[0]}' --soft-filter '{params.AF_max[1]}' --mode '+' | \
 bcftools filter --exclude 'INFO/STATUS ~ "germline/i"' --soft-filter '{params.possible_germline}' --mode '+' | \
 bcftools view -f PASS -o {output.vcf_filtered} -O z;
 

diff --git a/CHANGELOG.rst b/CHANGELOG.rst
@@ -101,6 +101,7 @@ Removed:
 * Realignment step for TGA workflow https://github.com/Clinical-Genomics/BALSAMIC/pull/1272
 * Archived/outdated workflows and scripts https://github.com/Clinical-Genomics/BALSAMIC/pull/1296
 * Sed command to convert CNVpytor integer to float, deprecated by updated CNVpytor version https://github.com/Clinical-Genomics/BALSAMIC/pull/1310
+* Removed max AF 1 filter from bcftools https://github.com/Clinical-Genomics/BALSAMIC/pull/1338
 * Extra samtools sort command from WGS cases https://github.com/Clinical-Genomics/BALSAMIC/pull/1334
 
 [12.0.2]