Merge pull request #598 from nf-core/deepvar_udpate

Add option to supply bed file with PAR regions & enable haploid analysis for males
nf-core · Aug 15, 2024 · b712e0c · b712e0c
2 parents 5ce1f18 + 24b2b42
commit b712e0c
Show file tree

Hide file tree

Showing 14 changed files with 189 additions and 4 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -7,6 +7,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### `Added`
 
+- A new parameter `par_bed` to pass a PAR bed files to deepvariant [#598](https://github.com/nf-core/raredisease/pull/598)
 - A new functionality to pass gzipped resources to vcfanno_extra_resources [#589](https://github.com/nf-core/raredisease/pull/589)
 - A new parameter `vcfanno_extra_resources` to pass an extra resource to vcfanno [#588](https://github.com/nf-core/raredisease/pull/588)
 - A new parameter `scatter_count` to control how many interval files are created from a genome (used to parallelize annotations) [#585](https://github.com/nf-core/raredisease/pull/585)
@@ -19,6 +20,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### `Changed`
 
+- Males' X and Y chromosomes will be treated as haploids during variant calling by deepvariant [#598](https://github.com/nf-core/raredisease/pull/598)
 - Acceptable type for lane field in the samplesheet from number to string [#597](https://github.com/nf-core/raredisease/pull/597)
 - Allow `0` as a valid value for `sex` in the samplesheet [#595](https://github.com/nf-core/raredisease/pull/595)
 - Updated deepvariant to version 1.6.1 [#587](https://github.com/nf-core/raredisease/pull/587)
@@ -57,6 +59,12 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 |                 | scatter_count            |
 |                 | vcfanno_extra_resources  |
 
+### Tool updates
+
+| Tool        | Old version | New version |
+| ----------- | ----------- | ----------- |
+| Deepvariant | 1.5.0       | 1.6.1       |
+
 ## 2.1.0 - Obelix [2024-05-29]
 
 ### `Added`

diff --git a/conf/modules/call_snv_deepvariant.config b/conf/modules/call_snv_deepvariant.config
@@ -22,7 +22,10 @@ process {
     }
 
     withName: '.*CALL_SNV_DEEPVARIANT:DEEPVARIANT' {
-        ext.args = { "--model_type=${params.analysis_type.toUpperCase()}" }
+        ext.args = { [
+            "--model_type=${params.analysis_type.toUpperCase()}",
+            meta.sex == "1" ? params.genome == 'GRCh37' ? '--haploid_contigs="X,Y"' : '--haploid_contigs="chrX,chrY"' : ''
+        ].join(' ') }
         ext.prefix = { "${meta.id}_deepvar" }
     }
 

diff --git a/docs/usage.md b/docs/usage.md
@@ -201,10 +201,12 @@ The mandatory and optional parameters for each category are tabulated below.
 | ml_model<sup>2</sup>       | known_dbsnp_tbi<sup>2</sup> |
 | analysis_type<sup>3</sup>  | call_interval<sup>2</sup>   |
 |                            | known_dbsnp_tbi<sup>2</sup> |
+|                            | par_bed<sup>4</sup>         |
 
 <sup>1</sup>Default variant caller is DeepVariant, but you have the option to use Sentieon as well.<br />
 <sup>2</sup>These parameters are only used by Sentieon.<br />
 <sup>3</sup>Default is WGS, but you have the option to choose WES as well.<br />
+<sup>4</sup>This parameter is only used by Deepvariant.<br />
 
 ##### 5. Variant calling - Structural variants
 

diff --git a/main.nf b/main.nf
@@ -39,6 +39,7 @@ params.mobile_element_references       = getGenomeAttribute('mobile_element_refe
 params.mobile_element_svdb_annotations = getGenomeAttribute('mobile_element_svdb_annotations')
 params.ml_model                        = getGenomeAttribute('ml_model')
 params.mt_fasta                        = getGenomeAttribute('mt_fasta')
+params.par_bed                         = getGenomeAttribute('par_bed')
 params.ploidy_model                    = getGenomeAttribute('ploidy_model')
 params.reduced_penetrance              = getGenomeAttribute('reduced_penetrance')
 params.readcount_intervals             = getGenomeAttribute('readcount_intervals')

diff --git a/modules.json b/modules.json
@@ -113,7 +113,7 @@
                     },
                     "deepvariant": {
                         "branch": "master",
-                        "git_sha": "ccf06a896339d2a4ed8594daa9f9adb827cb8189",
+                        "git_sha": "a7e8b8afd4fa82f20d745fa778bfdbf39c1f7efb",
                         "installed_by": ["modules"]
                     },
                     "eklipse": {

diff --git a/modules/nf-core/deepvariant/main.nf b/modules/nf-core/deepvariant/main.nf
diff --git a/modules/nf-core/deepvariant/meta.yml b/modules/nf-core/deepvariant/meta.yml
diff --git a/modules/nf-core/deepvariant/tests/main.nf.test b/modules/nf-core/deepvariant/tests/main.nf.test
diff --git a/modules/nf-core/deepvariant/tests/main.nf.test.snap b/modules/nf-core/deepvariant/tests/main.nf.test.snap
diff --git a/modules/nf-core/deepvariant/tests/nextflow-non-autosomal-calling.config b/modules/nf-core/deepvariant/tests/nextflow-non-autosomal-calling.config
diff --git a/nextflow_schema.json b/nextflow_schema.json
@@ -267,6 +267,14 @@
                     "description": "Path to mitochondrial FASTA genome file.",
                     "fa_icon": "fas fa-file"
                 },
+                "par_bed": {
+                    "type": "string",
+                    "exists": true,
+                    "format": "path",
+                    "fa_icon": "fas fa-file",
+                    "pattern": "^\\S+\\.bed(\\.gz)?$",
+                    "description": "Path to a BED file containing PAR regions (used by deepvariant)."
+                },
                 "ploidy_model": {
                     "type": "string",
                     "exists": true,

diff --git a/subworkflows/local/call_snv.nf b/subworkflows/local/call_snv.nf
@@ -28,6 +28,7 @@ workflow CALL_SNV {
         ch_dbsnp_tbi          // channel: [optional] [ val(meta), path(tbi) ]
         ch_call_interval      // channel: [mandatory] [ path(intervals) ]
         ch_ml_model           // channel: [mandatory] [ path(model) ]
+        ch_par_bed            // channel: [optional] [ val(meta), path(bed) ]
         ch_case_info          // channel: [mandatory] [ val(case_info) ]
         ch_foundin_header     // channel: [mandatory] [ path(header) ]
         ch_pcr_indel_model    // channel: [optional] [ val(sentieon_dnascope_pcr_indel_model) ]
@@ -50,6 +51,7 @@ workflow CALL_SNV {
                 ch_genome_bam_bai,
                 ch_genome_fasta,
                 ch_genome_fai,
+                ch_par_bed,
                 ch_case_info,
                 ch_foundin_header,
                 ch_genome_chrsizes

diff --git a/subworkflows/local/variant_calling/call_snv_deepvariant.nf b/subworkflows/local/variant_calling/call_snv_deepvariant.nf
@@ -16,6 +16,7 @@ workflow CALL_SNV_DEEPVARIANT {
         ch_bam_bai         // channel: [mandatory] [ val(meta), path(bam), path(bai) ]
         ch_genome_fasta    // channel: [mandatory] [ val(meta), path(fasta) ]
         ch_genome_fai      // channel: [mandatory] [ val(meta), path(fai) ]
+        ch_par_bed         // channel: [optional] [ val(meta), path(bed) ]
         ch_case_info       // channel: [mandatory] [ val(case_info) ]
         ch_foundin_header  // channel: [mandatory] [ path(header) ]
         ch_genome_chrsizes // channel: [mandatory] [ path(chrsizes) ]
@@ -28,7 +29,7 @@ workflow CALL_SNV_DEEPVARIANT {
             }
             .set { ch_deepvar_in }
 
-        DEEPVARIANT ( ch_deepvar_in, ch_genome_fasta, ch_genome_fai, [[],[]] )
+        DEEPVARIANT ( ch_deepvar_in, ch_genome_fasta, ch_genome_fai, [[],[]], ch_par_bed )
         DEEPVARIANT.out.gvcf
             .collect{it[1]}
             .toList()

diff --git a/workflows/raredisease.nf b/workflows/raredisease.nf
@@ -279,6 +279,8 @@ workflow RAREDISEASE {
     ch_mtshift_fai              = ch_references.mtshift_fai
     ch_mtshift_fasta            = ch_references.mtshift_fasta
     ch_mtshift_intervals        = ch_references.mtshift_intervals
+    ch_par_bed                  = params.par_bed                            ? Channel.fromPath(params.par_bed).map{ it -> [[id:'par_bed'], it] }.collect()
+                                                                            : Channel.value([[],[]])
     ch_ploidy_model             = params.ploidy_model                       ? Channel.fromPath(params.ploidy_model).map{ it -> [[id:it[0].simpleName], it] }.collect()
                                                                             : Channel.empty()
     ch_readcount_intervals      = params.readcount_intervals                ? Channel.fromPath(params.readcount_intervals).collect()
@@ -502,6 +504,7 @@ workflow RAREDISEASE {
             ch_dbsnp_tbi,
             ch_call_interval,
             ch_ml_model,
+            ch_par_bed,
             ch_case_info,
             ch_foundin_header,
             Channel.value(params.sentieon_dnascope_pcr_indel_model)