changes on giab benchmarking

Former-commit-id: 5fb9639
igsr · Feb 20, 2019 · 524cca8 · 524cca8
1 parent 20aa286
commit 524cca8
Show file tree

Hide file tree

Showing 5 changed files with 11 additions and 13 deletions.
diff --git a/scripts/VCF/FILTER/MLfilter_applymodel.nf b/scripts/VCF/FILTER/MLfilter_applymodel.nf
@@ -31,7 +31,6 @@ if (params.help) {
     log.info '  --annotations ANNOTATION_STRING String containing the annotations to filter, for example:'
     log.info '    %CHROM\t%POS\t%INFO/DP\t%INFO/RPB\t%INFO/MQB\t%INFO/BQB\t%INFO/MQSB\t%INFO/SGB\t%INFO/MQ0F\t%INFO/ICB\t%INFO/HOB\t%INFO/MQ\n.'
     log.info '  --vt  VARIANT_TYPE   Type of variant to filter. Poss1ible values are 'snps'/'indels'.'
-    log.info '  --split_multiallelics  true/false  If true then split the multiallelic positions. Default=false.'
     log.info '  --threads INT Number of threads used in the different BCFTools processes. Default=1.'
     log.info ''
     exit 1
@@ -45,8 +44,7 @@ chrChannel=Channel.from( chrList )
 
 //Apply a fitted model obtained after running MLfilter_trainmodel.nf
 
-if (params.split_multiallelics==true) {
-   process split_multiallelic {
+process split_multiallelic {
    	   /*
    	   This process is used to split the multiallelic sites into different lines per allele. It uses bcftools norm 
    	   for this
@@ -62,15 +60,15 @@ if (params.split_multiallelics==true) {
 
 	    output:
 	    file 'out.splitted.vcf.gz' into splitted_vcf
+	    file 'out.splitted.vcf.gz.tbi' into splitted_vcf_tbi
 	    val chr into chr1
 
 	    """
 	    bcftools norm -r ${chr} -m -${params.vt} ${params.vcf} -o out.splitted.vcf.gz -Oz --threads ${params.threads}
+	    tabix out.splitted.vcf.gz
 	    """
-    }
 }
 
-
 process get_variant_annotations {
 	/*
 	Process to get the variant annotations for the selected ${params.vt} from the unfiltered VCF file
@@ -83,13 +81,15 @@ process get_variant_annotations {
 
 	input:
 	val chr from chr1
+	file splitted_vcf
+	file splitted_vcf_tbi
 
 	output:
 	file 'unfilt_annotations.vt.tsv.gz' into unfilt_annotations
 	val chr into chr2
 
 	"""
-	bcftools view -c1 -r ${chr} -v ${params.vt} ${params.vcf} -o out.onlyvariants.vt.vcf.gz -Oz --threads ${params.threads}
+	bcftools view -c1 -r ${chr} -v ${params.vt} ${splitted_vcf} -o out.onlyvariants.vt.vcf.gz -Oz --threads ${params.threads}
 	tabix out.onlyvariants.vt.vcf.gz
 	bcftools query -H -r ${chr} -f '${params.annotations}' out.onlyvariants.vt.vcf.gz | bgzip -c > unfilt_annotations.vt.tsv.gz
 	"""
@@ -101,7 +101,7 @@ process apply_model {
 	and to apply this model on the unfiltered VCF
 	*/
 
-	memory '2 GB'
+	memory '5 GB'
         executor 'lsf'
         queue "${params.queue}"
         cpus 1

diff --git a/scripts/VCF/QC/compare_with_giab.config b/scripts/VCF/QC/compare_with_giab.config
@@ -1,4 +1,4 @@
-params.bcftools_folder='/homes/ernesto/bin/bcftools-1.6/' // folder containin the bcftools binary
+params.bcftools_folder='/nfs/production/reseq-info/work/bin/bcftools-1.9/' // folder containin the bcftools binary
 params.tabix='/nfs/production/reseq-info/work/ernesto/bin/anaconda3/bin/tabix' // path to tabix binary
 params.bgzip='/nfs/production/reseq-info/work/ernesto/bin/anaconda3/bin/bgzip' // path to bgzip
 params.non_valid_regions='/nfs/production/reseq-info/work/ernesto/isgr/SUPPORTING/REFERENCE/exclude_nonvalid.bed' // path to BED format file containing the regions that will be excluded from the comparison

diff --git a/scripts/VCF/QC/compare_with_giab.nf b/scripts/VCF/QC/compare_with_giab.nf
@@ -54,7 +54,7 @@ process excludeNonVariants {
         file 'out.sites.vcf.gz' into out_sites_vcf
 
 	"""
-	${params.bcftools_folder}/bcftools view -G -m2 -M2 -c1 ${params.vcf} -f.,PASS -r ${params.chros} -o out.sites.vcf.gz -Oz
+	${params.bcftools_folder}/bcftools view -m2 -M2 -c1 ${params.vcf} -f.,PASS -r ${params.chros} -o out.sites.vcf.gz -Oz
 	${params.tabix} out.sites.vcf.gz
 	"""
 }

diff --git a/scripts/VCF/UTILS/split_giab_into_chros.config b/scripts/VCF/UTILS/split_giab_into_chros.config
@@ -1,4 +1,2 @@
-params.bcftools_folder='/homes/ernesto/bin/bcftools-1.6/' // folder containin the bcftools binary
 params.tabix_folder='/nfs/production/reseq-info/work/ernesto/bin/anaconda3/bin/' // path to tabix binary
-params.giab='/nfs/production/reseq-info/work/ernesto/isgr/SUPPORTING/REFERENCE/GIAB/ANALYSIS_READY/CHR20/NA12878.giab.SNP.chr20.non_valid.reheaded.vcf.gz' 
-params.queue='production-rh7' // queue name used by LSF
+params.queue='production-rh74' // queue name used by LSF
diff --git a/scripts/VCF/UTILS/split_giab_into_chros.nf b/scripts/VCF/UTILS/split_giab_into_chros.nf
@@ -64,7 +64,7 @@ process splitVCF {
 	file "${params.prefix}.${chr}.biallelic.vcf.gz*" into chr_vcf
 
 	"""
-	${params.bcftools_folder}/bcftools view -m2 -M2 -r ${chr} ${params.giab_vcf} -o ${params.prefix}.${chr}.biallelic.vcf.gz -O z
+	bcftools view -m2 -M2 -r ${chr} ${params.giab_vcf} -o ${params.prefix}.${chr}.biallelic.vcf.gz -O z
 	${params.tabix_folder}/tabix ${params.prefix}.${chr}.biallelic.vcf.gz
 	"""
 }