Fix gatk and bcftools rule setup

moiexpositoalonsolab · Feb 11, 2025 · 9d7c550 · 9d7c550
1 parent cacf59b
commit 9d7c550
Show file tree

Hide file tree

Showing 4 changed files with 8 additions and 9 deletions.
diff --git a/workflow/envs/gatk.yaml b/workflow/envs/gatk.yaml
@@ -9,7 +9,7 @@ dependencies:
   - pandas ==2.2.2
   - numpy  ==2.0.0
   # Tools
-  - gatk4 ==4.5.0.0
+  - gatk4 ==4.6.1.0
   - snakemake-wrapper-utils ==0.6.2
   # For plotting the VSQR output, we need R
   - r #==4.1

diff --git a/workflow/rules/calling-bcftools.smk b/workflow/rules/calling-bcftools.smk
@@ -76,7 +76,7 @@ if config["settings"].get("contig-group-size"):
                 if platform.system() == "Darwin"
                 else ""
             ),
-            java_opts=config["params"]["picard"]["SortVcf-java-opts"] + " -Xmx" + config["params"]["picard"].get("SortVcf-mem-mb", 1024) + "m",
+            java_opts=config["params"]["picard"]["SortVcf-java-opts"] + " -Xmx" + str(config["params"]["picard"].get("SortVcf-mem-mb", 1024)) + "m",
         log:
             "logs/calling/picard/sort-genotyped.log",
         benchmark:

diff --git a/workflow/rules/calling-haplotypecaller.smk b/workflow/rules/calling-haplotypecaller.smk
@@ -13,8 +13,8 @@ def get_gatk_call_variants_params(wildcards, input):
         get_gatk_regions_param(
             regions=input.regions, default="--intervals '{}'".format(wildcards.contig)
         )
-        + " --native-pair-hmm-threads "
-        + str(config["params"]["gatk"]["HaplotypeCaller-threads"])
+        # + " --native-pair-hmm-threads "
+        # + str(config["params"]["gatk"]["HaplotypeCaller-threads"])
         + " "
         + config["params"]["gatk"]["HaplotypeCaller-extra"]
     )
@@ -68,9 +68,6 @@ rule call_variants:
         "benchmarks/calling/called/gatk-haplotypecaller/{sample}.{contig}.log"
     # Need to set threads here so that snakemake can plan the job scheduling properly
     threads: config["params"]["gatk"]["HaplotypeCaller-threads"]
-    # resources:
-    # Increase time limit in factors of 24h, if the job fails due to time limit.
-    # time = lambda wildcards, input, threads, attempt: int(1440 * int(attempt))
     params:
         # The function here is where the contig variable is propagated to haplotypecaller.
         # Took me a while to figure this one out...
@@ -79,6 +76,8 @@ rule call_variants:
         java_opts=config["params"]["gatk"]["HaplotypeCaller-java-opts"],
     resources:
         mem_mb=config["params"]["gatk"].get("HaplotypeCaller-mem-mb", 1024),
+        # Increase time limit in factors of 24h, if the job fails due to time limit.
+        # time = lambda wildcards, input, threads, attempt: int(1440 * int(attempt))
     group:
         "call_variants"
     conda:

diff --git a/workflow/rules/filtering-gatk-vqsr.smk b/workflow/rules/filtering-gatk-vqsr.smk
@@ -100,7 +100,7 @@ rule gatk_variant_recalibrator:
         annotation=config["params"]["gatk-vqsr"]["annotation"],
         # Extras
         extra=get_variant_recalibrator_extra,
-        java_opts=config["params"]["gatk-vqsr"]["variantrecalibrator-java-opts"] + " -Xmx" + config["params"]["gatk-vqsr"].get("variantrecalibrator-mem-mb", 1024) + "m",
+        java_opts=config["params"]["gatk-vqsr"]["variantrecalibrator-java-opts"] + " -Xmx" + str(config["params"]["gatk-vqsr"].get("variantrecalibrator-mem-mb", 1024)) + "m",
     log:
         "logs/calling/gatk-variantrecalibrator/{vartype}.log",
     benchmark:
@@ -147,7 +147,7 @@ rule gatk_apply_vqsr:
         # set mode, must be either SNP, INDEL or BOTH
         mode="{vartype}",
         extra=get_apply_vqsr_extra,
-        java_opts=config["params"]["gatk-vqsr"]["applyvqsr-java-opts"] + " -Xmx" + config["params"]["gatk-vqsr"].get("applyvqsr-mem-mb", 1024) + "m",
+        java_opts=config["params"]["gatk-vqsr"]["applyvqsr-java-opts"] + " -Xmx" + str(config["params"]["gatk-vqsr"].get("applyvqsr-mem-mb", 1024)) + "m",
     # resources:
     # mem_mb=50
     conda: