Add Expansionhunter

CHANGELOG.md

@@ -7,6 +7,13 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## dev
 
+### New features
+
+- Added `expansionhunter` as a new caller. This caller is unique as no postprocessing is done on the outputs of the caller.
+- Added `sex` as a samplesheet field. This is an optional field to improve the repeat expansion calls from Expansionhunter.
+
+### Fixes
+
 - Fixes a small issue with the VEP config (`.split()` error) by replacing it with the `name` attribute.
 
 ## v1.13.3

assets/schema_input.json

@@ -94,6 +94,12 @@
                 "type": "boolean",
                 "default": false,
                 "meta": "msi"
+            },
+            "sex": {
+                "type": "string",
+                "default": null,
+                "meta": "sex",
+                "enum": ["male", "female"]
             }
         },
         "anyOf": [

conf/modules.config

@@ -321,6 +321,21 @@ process {
         ext.prefix = {"${meta.sample}.${meta.roc_type}"}
     }
 
+    /*
+    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+        REPEATS DETECTION
+    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+    */
+
+    withName: ".*CRAM_REPEAT_EXPANSIONHUNTER:EXPANSIONHUNTER\$" {
+        ext.args = { meta.sex ? "--sex ${meta.sex}" : '' }
+    }
+
+    withName: ".*CRAM_REPEAT_EXPANSIONHUNTER:BCFTOOLS_ANNOTATE\$" {
+        ext.args = "-c INFO/REPREF:=INFO/REF --output-type z --write-index=tbi"
+        ext.prefix = { "${meta.id}.expansionhunter" }
+    }
+
     /*
     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
         FINAL PROCESSES

conf/test.config

@@ -38,6 +38,8 @@ params {
     strtablefile    = "https://github.com/nf-cmgg/test-datasets/raw/smallvariants/data/genomics/homo_sapiens/genome/hg38_chr21_22000000_23000000.strtable.zip"
     ped             = "https://github.com/nf-cmgg/test-datasets/raw/smallvariants/data/genomics/homo_sapiens/genome/test_dots.ped"
     msi_baseline    = "https://github.com/nf-cmgg/test-datasets/raw/smallvariants/data/genomics/homo_sapiens/genome/msi_subset.baseline.list"
+    somalier_sites  = "https://github.com/brentp/somalier/files/3412456/sites.hg38.vcf.gz"
+    expansionhunter_catalogue = "https://github.com/nf-cmgg/test-datasets/raw/refs/heads/smallvariants/data/genomics/homo_sapiens/genome/variant_catalog.json"
 
     // Pipeline specific parameters
     filter             = true

docs/usage.md

@@ -91,6 +91,7 @@ The samplesheet can have following columns:
 | `roi`            | OPTIONAL - Full path to a BED file containing the regions of interest for the current sample to call on. When this file is given, the pipeline will run this sample in WES mode. (The flag `--roi <path>` can also be given to run WES mode for all samples using the file specified by the flag) File has to have the extension `.bed` or `.bed.gz`.                 |
 | `vardict_min_af` | OPTIONAL - The minimum AF value to use for the vardict variant caller (`--callers vardict`). This can be set in the samplesheet when it differs for all samples. A default can be set using the `--vardict_min_af` parameter (whichs defaults to 0.1)                                                                                                                 |
 | `msi`            | OPTIONAL - Set this to `true` when MSI calling should be performed on this sample. MSI calling is only done when a CRAM file is supplied and a baseline file has been provided via the `--msi_baseline` parameter. (By default this option is `false`)                                                                                                                |
+| `sex`            | OPTIONAL - The sex of the sample. This is an optional field used by expansionhunter to improve the repeat expansion calls.                                                                                                                                                                                                                                            |
 
 <!-- prettier-ignore -->
 !!!note

main.nf

@@ -67,6 +67,9 @@ params {
     // Path to the elsites file. This is used when `elprep` is part of the callers.
     elsites: Path?
 
+    // Path to the ExpansionHunter variant catalogue JSON file. This is needed when `expansionhunter` is part of the callers.
+    expansionhunter_catalogue: Path? = getGenomeAttribute('expansionhunter_catalogue', params.genomes, params.genome)
+
     // Object for genomes
     genomes: Map = [:]
 
@@ -351,7 +354,7 @@ workflow {
     // Check for dependencies between parameters
     //
 
-    def List<String> available_callers = ["haplotypecaller", "vardict", "elprep"]
+    def List<String> available_callers = ["haplotypecaller", "vardict", "elprep", "expansionhunter"]
 
     if(params.dbsnp_tbi && !params.dbsnp){
         error("Please specify the dbsnp VCF with --dbsnp VCF")
@@ -379,6 +382,10 @@ workflow {
         if(!(caller in available_callers)) { error("\"${caller}\" is not a supported callers please use one or more of these instead: ${available_callers.join(', ')}") }
     }
 
+    if (callers.contains("expansionhunter") && !params.expansionhunter_catalogue) {
+        error("Please specify the ExpansionHunter variant catalogue JSON file with --expansionhunter_catalogue")
+    }
+
     /*
     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
         CONFIG FILES
@@ -450,6 +457,7 @@ workflow {
         params.elsites,
         params.msi_baseline,
         params.updio_regions,
+        params.expansionhunter_catalogue,
 
         // Boolean inputs
         params.dragstr,
@@ -502,6 +510,7 @@ workflow {
     gvcf_reports        = SMALLVARIANTS.out.gvcf_reports
     genomicsdb          = SMALLVARIANTS.out.genomicsdb
     vcfs                = SMALLVARIANTS.out.vcfs.filter { _meta, vcf, _tbi -> vcf.startsWith(workflow.workDir) } // Filtering out input VCFs from the output publishing fixes an issue in the current implementation of the workflow output definitions: https://github.com/nextflow-io/nextflow/issues/5480
+    repeat_vcfs         = SMALLVARIANTS.out.repeat_vcfs
     gemini              = SMALLVARIANTS.out.gemini
     peds                = SMALLVARIANTS.out.peds
     joint_beds          = SMALLVARIANTS.out.joint_beds
@@ -553,6 +562,10 @@ output {
         vcf >> "${meta.family}/output_${params.unique_out}/${meta.id}.${meta.caller}.vcf.gz"
         tbi >> "${meta.family}/output_${params.unique_out}/${meta.id}.${meta.caller}.vcf.gz.tbi"
     } }
+    repeat_vcfs { path { meta, vcf, tbi ->
+        vcf >> "${meta.family}/output_${params.unique_out}/${meta.id}.${meta.caller}.vcf.gz"
+        tbi >> "${meta.family}/output_${params.unique_out}/${meta.id}.${meta.caller}.vcf.gz.tbi"
+    } }
     gemini { path { meta, db ->
         db >> "${meta.family}/output_${params.unique_out}/${meta.id}.${meta.caller}.db"
     } }

modules.json

@@ -87,6 +87,11 @@
                         "git_sha": "6d46786420b4d7bc88eba026eb389c0c5535d120",
                         "installed_by": ["modules"]
                     },
+                    "expansionhunter": {
+                        "branch": "master",
+                        "git_sha": "0adfbfad03ff9af0d9fb437f0f2e1140e1c59316",
+                        "installed_by": ["modules"]
+                    },
                     "gatk4/calibratedragstrmodel": {
                         "branch": "master",
                         "git_sha": "6d46786420b4d7bc88eba026eb389c0c5535d120",