move to annovar because of to many snpEff bugs...

tbersez · Apr 16, 2019 · a16323d · a16323d
1 parent b4f7fcf
commit a16323d
Show file tree

Hide file tree

Showing 4 changed files with 85 additions and 10 deletions.
diff --git a/Snakefile b/Snakefile
@@ -39,17 +39,14 @@ include : cwd + "modules/" + config["FASTP"]
 include : cwd + "modules/" + config["INDEXER"]
 include : cwd + "modules/" + config["ALLIGNER"]
 include : cwd + "modules/bam_with_bed_parse.py"
+include : cwd + "modules/vcf_to_avinput.py"
 include : cwd + "modules/varscan.py"
-include : cwd + "modules/snpEff.py"
-include : cwd + "modules/snpsift_non_synonymous.py"
+include : cwd + "modules/annovar.py"
 include : cwd + "modules/make_report.py"
 include : cwd + "modules/bam_index.py"
 include : cwd + "modules/whatshap.py"
-include : cwd + "modules/vcf_merge.py"
 
 # target files...
 rule all:
     input:
-        'Non_synonymous_variants_summary.tab',
-        expand(config["VAR"] + "{samples}/{samples}_varscan_phased.vcf", samples = config["samples"]),
-        config["VAR"] + "Non_synonymous_SNP.vcf"
+        expand(config["VAR"] + "{samples}/{samples}_varscan.avinput.variant_function", samples = config["samples"])
diff --git a/modules/annovar.py b/modules/annovar.py
@@ -0,0 +1,34 @@
+# Annovar SNPs annotation
+#
+#   This module annotate SNPs from Varscan
+#   on a gene model basis
+#
+#   Input:
+#       - sample_varscan.avinput
+#
+#   Output:
+#       - sample_varscan_anno.avoutput
+#
+#   Parameters:
+#       None...
+
+rule annovar :
+    input:
+        inav = config["VAR"] + "{samples}/{samples}_varscan.avinput"
+    output:
+        all_vars = config["VAR"] + "{samples}/{samples}_varscan.avinput.variant_function",
+        exonic_vars = config["VAR"] + "{samples}/{samples}_varscan.avinput.exonic_variant_function"
+    params:
+        bind = config["BIND"],
+        cont = config["CONT"]
+    message: "Annotating {input.inav} with Annovar \n"
+    shell:
+        """
+        singularity exec -B {params.bind} {params.cont} \
+        /opt/annovar/annotate_variation.pl \
+        -geneanno \
+        -dbtype refGene \
+        -buildver AV \
+        {input.inav} \
+        avdb/ \
+        """
diff --git a/modules/vcf_to_avinput.py b/modules/vcf_to_avinput.py
@@ -0,0 +1,32 @@
+# Annovar format converter
+#
+#   This module convert vcf4 format files from Varscan
+#   to annovar input format
+#
+#   Input:
+#       - sample(vcf4)_varscan.vcf
+#
+#   Output:
+#       - sample_varscan.avinput
+#
+#   Parameters:
+#       None...
+
+rule vcf4_to_avinput:
+    input:
+        vcf = config["VAR"] + "{samples}/{samples}_varscan.vcf"
+    output:
+        annovar = config["VAR"] + "{samples}/{samples}_varscan.avinput"
+    params:
+        bind = config["BIND"],
+        cont = config["CONT"]
+    message: "Converting {input.vcf} to avinput format \n"
+    shell:
+        """
+        singularity exec -B {params.bind} {params.cont} \
+        /opt/annovar/convert2annovar.pl \
+        -format vcf4 \
+        -includeinfo \
+        -outfile {output.annovar} \
+        {input.vcf}
+        """
diff --git a/slurm_config.json b/slurm_config.json
@@ -1,13 +1,15 @@
 {
     "__default__" : {
       "name"      : "AllMine {rule} [{wildcards}]",
-      "partition"      : "express",
-      "time"           : "48:00:00",
+      "partition"      : "lm.q",
+      "time"           : "72:00:00",
       "ntasks"         : 1,
       "cpus-per-task"  : 1,
       "mem-per-cpu-mb" : 2000,
-      "output"    : "~/logs/{rule}.{wildcards}.stout",
-      "error"     : "~/logs/{rule}.{wildcards}.sterr"
+      "errorStrategy" : "retry",
+      "maxRetries" : 2,
+      "cache" : "deep",
+      "maxErrors" : 10
     },
 
     "fastp_paired" : {
@@ -111,5 +113,15 @@
     "bam_index" : {
       "cpus-per-task"  : 1,
       "mem-per-cpu-mb" : 2000
+    },
+
+    "annovar" : {
+      "cpus-per-task"  : 1,
+      "mem-per-cpu-mb" : 6000
+    },
+
+    "vcf4_to_avinput" : {
+      "cpus-per-task"  : 1,
+      "mem-per-cpu-mb" : 4000
     }
 }