Genmod fixes

configs/nextflow.hopper.config

@@ -88,9 +88,14 @@ params {
 
   // CONTAINERS //
   container_vep       = "/fs1/resources/containers/ensembl-vep_release_111.0.sif"
-  container_fastp     = '/fs1/resources/containers/fastp_0.23.4.sif'
-  container_sentieon  = '/fs1/resources/containers/sentieon_202308.sif'
+  container_fastp     = "/fs1/resources/containers/fastp_0.23.4.sif"
+  container_sentieon  = "/fs1/resources/containers/sentieon_202308.sif"
   container_d4tools   = "/fs1/resources/containers/d4tools_0.3.8.sif"
+  container_genmod    = "/fs1/resources/containers/genmod_v3.9.sif"
+  container_bcftools  = "/fs1/resources/containers/bcftools_1.20.sif"
+
+  genmod_compound_trio_threshold    = 12
+  genmod_compound_trio_penalty      = 8
 
   run_chanjo2       = true
   reanalyze         = false

docs/list_of_all_used_software.md

@@ -13,7 +13,7 @@
 | fastp                  | 0.23.4       | Open-Source | https://github.com/OpenGene/fastp                      | https://github.com/OpenGene/fastp/issues               | Sailendra Pradhananga | FASTQ preprocessor                                                                        |
 | freebayes              | 1.3.2        | Open-Source | https://github.com/freebayes/freebayes                 | https://github.com/freebayes/freebayes/issues          | Ram Sai Nanduri       | Bayesian haplotype-based genetic polymorphism discovery and genotyping                    |
 | gatk                   | 4.1.9.0      | Open-Source | https://gatk.broadinstitute.org/hc/en-us               | https://github.com/broadinstitute/gatk/issues          | Viktor Henmyr         | Wide set of tools for genomic analysis                                                    |
-| genmod                 | 3.7.4        | Open-Source | https://github.com/Clinical-Genomics/genmod            | https://github.com/Clinical-Genomics/genmod/issues     | Viktor Henmyr         | Annotate models of genetic inheritance patterns in variant files                          |
+| genmod                 | 3.9.0        | Open-Source | https://github.com/Clinical-Genomics/genmod            | https://github.com/Clinical-Genomics/genmod/issues     | Viktor Henmyr         | Annotate models of genetic inheritance patterns in variant files                          |
 | haplogrep              | 2.2.8        | Open-Source | https://github.com/seppinho/haplogrep-cmd              | https://github.com/seppinho/haplogrep-cmd/issues       | Paul Piccinelli       | mtDNA haplogroup classification. Supporting rCRS and RSRS.                                |
 | hmtnote                | 0.7.2        | Open-Source | https://github.com/robertopreste/HmtNote               | https://github.com/robertopreste/HmtNote/issues        | Paul Piccinelli       | Human mitochondrial variants annotation using HmtVar                                      |
 | madeleine              | 2.0          | Open-Source | https://github.com/piratical/Madeline_2.0_PDE          | https://github.com/piratical/Madeline_2.0_PDE/issues   | Jakob Willforss       | Pedigree drawing program                                                                  |

main.nf

@@ -2175,27 +2175,60 @@ def calculate_indel_cadd_version(task) {
 	"""
 }
 
+process bgzip_indel_cadd {
+	cpus 4
+	tag "$group"
+	memory '1 GB'
+	time '5m'
+	container = "${params.container_bcftools}"
+
+	input:
+		set group, file(cadd_scores) from indel_cadd
+
+	output:
+		set group, file("cadd.gz"), file("cadd.gz.tbi") into indel_cadd_bgzip
+
+	script:
+		"""
+		gunzip -c ${cadd_scores} > cadd
+		bgzip -@ ${task.cpus} cadd
+		tabix -p vcf cadd.gz
+		"""
+
+	stub:
+		"""
+		touch "${group}.cadd.FIXME"
+		${bgzip_indel_cadd_version(task)}
+		"""
+}
+def bgzip_indel_cadd_version(task) {
+	"""
+	cat <<-END_VERSIONS > ${task.process}_versions.yml
+	${task.process}:
+	    tabix: \$(echo \$(tabix --version 2>&1) | sed 's/^.*(htslib) // ; s/ Copyright.*//')
+	    bcftools: \$(echo \$(bcftools --version 2>&1) | head -n1 | sed 's/^.*bcftools //; s/ .*\$//')
+	END_VERSIONS
+	"""
+}
+
 // Add the calculated indel CADDs to the vcf
 process add_cadd_scores_to_vcf {
 	cpus 4
 	tag "$group"
 	memory '1 GB'
 	time '5m'
-	container = '/fs1/resources/containers/genmod.sif'
+	container = "${params.container_genmod}"
 
 	input: 
-		set group, file(vcf), file(cadd_scores) from splice_marked.join(indel_cadd)
+		set group, file(vcf), file(cadd_scores), file(cadd_scores_tbi) from splice_marked.join(indel_cadd_bgzip)
 
 	output:
 		set group, file("${group}.cadd.vcf") into ma_vcf, fa_vcf, base_vcf
 		set group, file("*versions.yml") into ch_add_cadd_scores_to_vcf_versions
 
 	script:
 		"""
-		gunzip -c $cadd_scores > cadd
-		bgzip -@ ${task.cpus} cadd
-		tabix -p vcf cadd.gz
-		genmod annotate --cadd-file cadd.gz $vcf > ${group}.cadd.vcf
+		genmod annotate --cadd-file ${cadd_scores} ${vcf} > ${group}.cadd.vcf
 
 		${add_cadd_scores_to_vcf_version(task)}
 		"""
@@ -2210,10 +2243,8 @@ def add_cadd_scores_to_vcf_version(task) {
 	"""
 	cat <<-END_VERSIONS > ${task.process}_versions.yml
 	${task.process}:
-	    tabix: \$(echo \$(tabix --version 2>&1) | sed 's/^.*(htslib) // ; s/ Copyright.*//')
 	    genmod: \$(echo \$(genmod --version 2>&1) | sed -e "s/^.*genmod version: //")
 	END_VERSIONS
-
 	"""
 }
 
@@ -2227,7 +2258,7 @@ process inher_models {
 	// scratch true
 	// stageInMode 'copy'
 	// stageOutMode 'copy'
-	container = '/fs1/resources/containers/genmod.sif'
+	container = "${params.container_genmod}"
 
 	input:
 		set group, file(vcf), type, file(ped) from base_vcf.mix(ma_vcf, fa_vcf).join(ped_inher.mix(ped_inher_ma,ped_inher_fa))
@@ -2266,7 +2297,7 @@ process genmodscore {
 	tag "$group"
 	memory '20 GB'
 	time '1h'
-	container = '/fs1/resources/containers/genmod.sif'
+	container = "${params.container_genmod}"
 
 	input:
 		set group, type, file(vcf) from inhermod
@@ -2284,7 +2315,11 @@ process genmodscore {
 		if ( mode == "family" && params.antype == "wgs" ) {
 			"""
 			genmod score -i $group_score -c $params.rank_model -r $vcf -o ${group_score}.score1.vcf
-			genmod compound ${group_score}.score1.vcf > ${group_score}.score2.vcf
+			genmod compound \
+				--threshold ${params.genmod_compound_trio_threshold} \
+				--penalty ${params.genmod_compound_trio_penalty} \
+				-o ${group_score}.score2.vcf \
+				${group_score}.score1.vcf
 			sed 's/RankScore=${group}:/RankScore=${group_score}:/g' -i ${group_score}.score2.vcf
 			genmod sort -p -f $group_score ${group_score}.score2.vcf -o ${group_score}.scored.vcf
 
@@ -2294,7 +2329,18 @@ process genmodscore {
 		else {
 			"""
 			genmod score -i $group_score -c $params.rank_model_s -r $vcf -o ${group_score}.score1.vcf
-			genmod sort -p -f $group_score ${group_score}.score1.vcf -o ${group_score}.scored.vcf
+
+			// To get compounds without applying rank score penalty
+			genmod compound \
+				--penalty 0 \
+				-o ${group_score}.score1.with_compounds.vcf \
+				${group_score}.score1.vcf
+
+			genmod sort \
+				-p \
+				-f $group_score \
+				-o ${group_score}.scored.vcf \
+				${group_score}.score1.with_compounds.vcf
 
 			${genmodscore_version(task)}
 			"""
@@ -3685,52 +3731,33 @@ process score_sv {
 	publishDir "${OUTDIR}/vcf", mode: 'copy', overwrite: 'true', pattern: '*.vcf.gz*'
 	memory '10 GB'
 	time '2h'
-	container = '/fs1/resources/containers/genmod.sif'
+	container = "${params.container_genmod}"
 
 	input:
 		set group, type, file(vcf) from annotatedSV
 
 	output:
-		set group, type, file("${group_score}.sv.scored.sorted.vcf.gz"), file("${group_score}.sv.scored.sorted.vcf.gz.tbi") into sv_rescore,sv_rescore_ma,sv_rescore_fa
-		set group, file("${group}_sv.INFO") into sv_INFO
-		set group, file("${group_score}.sv.scored.sorted.vcf.gz") into svvcf_bed, svvcf_pod
+		set group, val(group_score), file(".sv.scored.vcf") into ch_scored_sv
 		set group, file("*versions.yml") into ch_score_sv_versions
 
 	script:
-		group_score = group
-		if ( type == "ma" || type == "fa") {
+		def model = (mode == "family" && params.antype == "wgs") ? params.svrank_model : params.svrank_model_s
+		def scoredVcfOutput = (mode == "family" && params.antype == "wgs") ? "${group_score}.sv.scored.vcf" : "${group_score}.sv.scored.vcf"
+		def group_score = group
+		if ( type == "ma" || type == "fa" ) {
 			group_score = group + "_" + type
 		}
 
-		if (mode == "family" && params.antype == "wgs") {
-			"""
-			genmod score -i $group_score -c $params.svrank_model -r $vcf -o ${group_score}.sv.scored_tmp.vcf
-			bcftools sort -O v -o ${group_score}.sv.scored.sorted.vcf ${group_score}.sv.scored_tmp.vcf 
-			bgzip -@ ${task.cpus} ${group_score}.sv.scored.sorted.vcf -f
-			tabix ${group_score}.sv.scored.sorted.vcf.gz -f
-			echo "SV	$type	${params.accessdir}/vcf/${group_score}.sv.scored.sorted.vcf.gz" > ${group}_sv.INFO
-
-			${score_sv_version(task)}
-			"""
-		}
-		else {
-			"""
-			genmod score -i $group_score -c $params.svrank_model_s -r $vcf -o ${group_score}.sv.scored.vcf
-			bcftools sort -O v -o ${group_score}.sv.scored.sorted.vcf ${group}.sv.scored.vcf
-			bgzip -@ ${task.cpus} ${group_score}.sv.scored.sorted.vcf -f
-			tabix ${group_score}.sv.scored.sorted.vcf.gz -f
-			echo "SV	$type	${params.accessdir}/vcf/${group_score}.sv.scored.sorted.vcf.gz" > ${group}_sv.INFO
+		"""
+		genmod score -i ${group_score} -c ${model} -r ${vcf} -o ${group_score}.sv.scored.vcf
 
-			${score_sv_version(task)}
-			"""
-		}
+		${score_sv_version(task)}
+		"""
 
 	stub:
 		group_score = group
 		"""
-		touch "${group_score}.sv.scored.sorted.vcf.gz"
-		touch "${group_score}.sv.scored.sorted.vcf.gz.tbi"
-		touch "${group}_sv.INFO"
+		touch "${group_score}.sv.scored.sorted.vcf"
 
 		${score_sv_version(task)}
 		"""
@@ -3739,8 +3766,50 @@ def score_sv_version(task) {
 	"""
 	cat <<-END_VERSIONS > ${task.process}_versions.yml
 	${task.process}:
-	    tabix: \$(echo \$(tabix --version 2>&1) | sed 's/^.*(htslib) // ; s/ Copyright.*//')
 	    genmod: \$(echo \$(genmod --version 2>&1) | sed -e "s/^.*genmod version: //")
+	END_VERSIONS	
+	"""
+}
+
+process bgzip_scored_genmod {
+	cpus 4
+	tag "$group"
+	memory '1 GB'
+	time '5m'
+	container = "${params.container_bcftools}"
+
+	input:
+		set group, val(group_score), file(scored_sv_vcf) from ch_scored_sv
+
+	output:
+		set group, type, file("${group_score}.sv.scored.sorted.vcf.gz"), file("${group_score}.sv.scored.sorted.vcf.gz.tbi") into sv_rescore, sv_rescore_ma, sv_rescore_fa
+		set group, file("${group_score}.sv.scored.sorted.vcf.gz") into svvcf_bed, svvcf_pod
+		set group, file("${group}_sv.INFO") into sv_INFO
+		set group, file("*versions.yml") into ch_score_sv_bgzip_versions
+
+	script:
+		"""
+			bcftools sort -O v -o ${group_score}.sv.scored.sorted.vcf ${scored_sv_vcf}
+			bgzip -@ ${task.cpus} ${group_score}.sv.scored.sorted.vcf -f
+			tabix ${group_score}.sv.scored.sorted.vcf.gz -f
+			echo "SV\t$type\t${params.accessdir}/vcf/${group_score}.sv.scored.sorted.vcf.gz" > ${group}_sv.INFO
+
+			${bgzip_score_sv_version(task)}
+		"""
+	stub:
+		"""
+			touch "${group_score}.sv.scored.sorted.vcf.gz"
+			touch "${group_score}.sv.scored.sorted.vcf.gz.tbi"
+			touch "${group}_sv.INFO"
+
+			${bgzip_score_sv_version(task)}
+		"""
+}
+def bgzip_score_sv_version(task) {
+	"""
+	cat <<-END_VERSIONS > ${task.process}_versions.yml
+	${task.process}:
+	    tabix: \$(echo \$(tabix --version 2>&1) | sed 's/^.*(htslib) // ; s/ Copyright.*//')
 	    bcftools: \$(echo \$(bcftools --version 2>&1) | head -n1 | sed 's/^.*bcftools //; s/ .*\$//')
 	END_VERSIONS	
 	"""