add index for phase

wtsi-hgi · May 1, 2024 · 8877bf2 · 8877bf2
1 parent a1d3ba8
commit 8877bf2
Show file tree

Hide file tree

Showing 9 changed files with 298 additions and 8 deletions.
diff --git a/modules.json b/modules.json
@@ -5,6 +5,20 @@
         "https://github.com/nf-core/modules.git": {
             "modules": {
                 "nf-core": {
+                    "bcftools/index": {
+                        "branch": "master",
+                        "git_sha": "44096c08ffdbc694f5f92ae174ea0f7ba0f37e09",
+                        "installed_by": [
+                            "modules"
+                        ]
+                    },
+                    "bcftools/merge": {
+                        "branch": "master",
+                        "git_sha": "44096c08ffdbc694f5f92ae174ea0f7ba0f37e09",
+                        "installed_by": [
+                            "modules"
+                        ]
+                    },
                     "glimpse2/ligate": {
                         "branch": "master",
                         "git_sha": "ee7fee68281944b002bd27a8ff3f19200b4d3fad",

diff --git a/modules/local/split_vcfs/main.nf b/modules/local/split_vcfs/main.nf
@@ -3,15 +3,15 @@ process SPLIT_VCFS {
     label 'process_medium'
 
     input:
-    tuple path(vcf), path(sample_list)
+    tuple val(meta), path(vcf), path(sample_list)
 
     output:
-    tuple path('vcf_sample_subset_*.vcf.gz'), path('vcf_sample_subset_*.vcf.gz.csi'), emit: split_vcfs
+    tuple val(meta), path('vcf_sample_subset_*.vcf.gz'), path('vcf_sample_subset_*.vcf.gz.csi'), emit: split_vcfs
 
     script:
     """
-    bcftools view -S $sample_list $vcf -Oz -o vcf_sample_subset_${sample_list}.vcf.gz
-    bcftools index vcf_sample_subset_${sample_list}.vcf.gz
+    bcftools view -S $sample_list $vcf -Oz -o vcf_sample_subset_${meta.id}.vcf.gz
+    bcftools index vcf_sample_subset_${meta.id}.vcf.gz
     """
 
 }
diff --git a/modules/nf-core/bcftools/index/environment.yml b/modules/nf-core/bcftools/index/environment.yml
@@ -0,0 +1,7 @@
+name: bcftools_index
+channels:
+  - conda-forge
+  - bioconda
+  - defaults
+dependencies:
+  - bioconda::bcftools=1.18
diff --git a/modules/nf-core/bcftools/index/main.nf b/modules/nf-core/bcftools/index/main.nf
@@ -0,0 +1,51 @@
+process BCFTOOLS_INDEX {
+    tag "$meta.id"
+    label 'process_low'
+
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/bcftools:1.18--h8b25389_0':
+        'biocontainers/bcftools:1.18--h8b25389_0' }"
+
+    input:
+    tuple val(meta), path(vcf)
+
+    output:
+    tuple val(meta), path("*.csi"), optional:true, emit: csi
+    tuple val(meta), path("*.tbi"), optional:true, emit: tbi
+    path "versions.yml"           , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+
+    """
+    bcftools \\
+        index \\
+        $args \\
+        --threads $task.cpus \\
+        $vcf
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        bcftools: \$(bcftools --version 2>&1 | head -n1 | sed 's/^.*bcftools //; s/ .*\$//')
+    END_VERSIONS
+    """
+
+    stub:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    def extension = args.contains("--tsi") || args.contains("-t") ? "tbi" :
+                    "csi"
+    """
+    touch ${vcf}.${extension}
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        bcftools: \$(bcftools --version 2>&1 | head -n1 | sed 's/^.*bcftools //; s/ .*\$//')
+    END_VERSIONS
+    """
+}
diff --git a/modules/nf-core/bcftools/index/meta.yml b/modules/nf-core/bcftools/index/meta.yml
@@ -0,0 +1,48 @@
+name: bcftools_index
+description: Index VCF tools
+keywords:
+  - vcf
+  - index
+  - bcftools
+  - csi
+  - tbi
+tools:
+  - bcftools:
+      description: BCFtools is a set of utilities that manipulate variant calls in the Variant Call Format (VCF) and its binary counterpart BCF. All commands work transparently with both VCFs and BCFs, both uncompressed and BGZF-compressed.  Most commands accept VCF, bgzipped VCF and BCF with filetype detected automatically even when streaming from a pipe. Indexed VCF and BCF will work in all situations. Un-indexed VCF and BCF and streams will work in most, but not all situations.
+      homepage: https://samtools.github.io/bcftools/
+      documentation: https://samtools.github.io/bcftools/howtos/index.html
+      tool_dev_url: https://github.com/samtools/bcftools
+      doi: "10.1093/gigascience/giab008"
+      licence: ["MIT", "GPL-3.0-or-later"]
+input:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - VCF:
+      type: file
+      description: VCF file (optionally GZIPPED)
+      pattern: "*.{vcf,vcf.gz}"
+output:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - version:
+      type: file
+      description: File containing software version
+      pattern: "versions.yml"
+  - csi:
+      type: file
+      description: Default VCF file index file
+      pattern: "*.csi"
+  - tbi:
+      type: file
+      description: Alternative VCF file index file for larger files (activated with -t parameter)
+      pattern: "*.tbi"
+authors:
+  - "@jfy133"
+maintainers:
+  - "@jfy133"
diff --git a/modules/nf-core/bcftools/merge/environment.yml b/modules/nf-core/bcftools/merge/environment.yml
@@ -0,0 +1,7 @@
+name: bcftools_merge
+channels:
+  - conda-forge
+  - bioconda
+  - defaults
+dependencies:
+  - bioconda::bcftools=1.18
diff --git a/modules/nf-core/bcftools/merge/main.nf b/modules/nf-core/bcftools/merge/main.nf
@@ -0,0 +1,64 @@
+process BCFTOOLS_MERGE {
+    tag "$meta.id"
+    label 'process_medium'
+
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/bcftools:1.18--h8b25389_0':
+        'biocontainers/bcftools:1.18--h8b25389_0' }"
+
+    input:
+    tuple val(meta), path(vcfs), path(tbis)
+    tuple val(meta2), path(fasta)
+    tuple val(meta3), path(fai)
+    path(bed)
+
+    output:
+    tuple val(meta), path("*.{bcf,vcf}{,.gz}"), emit: merged_variants
+    path "versions.yml"                       , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args   ?: ''
+    def prefix   = task.ext.prefix ?: "${meta.id}"
+
+    def regions = bed ? "--regions-file $bed" : ""
+    def extension = args.contains("--output-type b") || args.contains("-Ob") ? "bcf.gz" :
+                    args.contains("--output-type u") || args.contains("-Ou") ? "bcf" :
+                    args.contains("--output-type z") || args.contains("-Oz") ? "vcf.gz" :
+                    args.contains("--output-type v") || args.contains("-Ov") ? "vcf" :
+                    "vcf"
+
+    """
+    bcftools merge \\
+        $args \\
+        $regions \\
+        --threads $task.cpus \\
+        --output ${prefix}.${extension} \\
+        $vcfs
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        bcftools: \$(bcftools --version 2>&1 | head -n1 | sed 's/^.*bcftools //; s/ .*\$//')
+    END_VERSIONS
+    """
+
+    stub:
+    def args = task.ext.args   ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    def extension = args.contains("--output-type b") || args.contains("-Ob") ? "bcf.gz" :
+                    args.contains("--output-type u") || args.contains("-Ou") ? "bcf" :
+                    args.contains("--output-type z") || args.contains("-Oz") ? "vcf.gz" :
+                    args.contains("--output-type v") || args.contains("-Ov") ? "vcf" :
+                    "vcf"
+    """
+    touch ${prefix}.${extension}
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        bcftools: \$(bcftools --version 2>&1 | head -n1 | sed 's/^.*bcftools //; s/ .*\$//')
+    END_VERSIONS
+    """
+}
diff --git a/modules/nf-core/bcftools/merge/meta.yml b/modules/nf-core/bcftools/merge/meta.yml
@@ -0,0 +1,88 @@
+name: bcftools_merge
+description: Merge VCF files
+keywords:
+  - variant calling
+  - merge
+  - VCF
+tools:
+  - merge:
+      description: |
+        Merge VCF files.
+      homepage: http://samtools.github.io/bcftools/bcftools.html
+      documentation: http://www.htslib.org/doc/bcftools.html
+      doi: 10.1093/bioinformatics/btp352
+      licence: ["MIT"]
+input:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - vcfs:
+      type: file
+      description: |
+        List containing 2 or more vcf files
+        e.g. [ 'file1.vcf', 'file2.vcf' ]
+  - tbis:
+      type: file
+      description: |
+        List containing the tbi index files corresponding to the vcfs input files
+        e.g. [ 'file1.vcf.tbi', 'file2.vcf.tbi' ]
+  - meta2:
+      type: map
+      description: |
+        Groovy Map containing reference information
+        e.g. [ id:'genome' ]
+  - fasta:
+      type: file
+      description: "(Optional) The fasta reference file (only necessary for the `--gvcf FILE` parameter)"
+      pattern: "*.{fasta,fa}"
+  - meta3:
+      type: map
+      description: |
+        Groovy Map containing reference information
+        e.g. [ id:'genome' ]
+  - fai:
+      type: file
+      description: "(Optional) The fasta reference file index (only necessary for the `--gvcf FILE` parameter)"
+      pattern: "*.fai"
+  - bed:
+      type: file
+      description: "(Optional) The bed regions to merge on"
+      pattern: "*.bed"
+output:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - vcf_gz:
+      type: file
+      description: VCF merged output file (bgzipped) => when `--output-type z` is used
+      pattern: "*.vcf.gz"
+  - vcf:
+      type: file
+      description: VCF merged output file => when `--output-type v` is used
+      pattern: "*.vcf"
+  - bcf_gz:
+      type: file
+      description: BCF merged output file (bgzipped) => when `--output-type b` is used
+      pattern: "*.bcf.gz"
+  - bcf:
+      type: file
+      description: BCF merged output file => when `--output-type u` is used
+      pattern: "*.bcf"
+  - versions:
+      type: file
+      description: File containing software versions
+      pattern: "versions.yml"
+authors:
+  - "@joseespinosa"
+  - "@drpatelh"
+  - "@nvnieuwk"
+  - "@ramprasadn"
+maintainers:
+  - "@joseespinosa"
+  - "@drpatelh"
+  - "@nvnieuwk"
+  - "@ramprasadn"
diff --git a/workflows/run_glimpse.nf b/workflows/run_glimpse.nf
@@ -1,6 +1,8 @@
 include { SPLIT_SAMPLES } from '../modules/local/split_samples/main'
 include { SPLIT_VCFS } from '../modules/local/split_vcfs/main'
 include { GLIMPSE2_PHASE } from '../modules/nf-core/glimpse2/phase/main'
+include { BCFTOOLS_INDEX as INDEX_PHASE  } from '../modules/nf-core/bcftools/index/main.nf'
+include { BCFTOOLS_INDEX as INDEX_LIGATE } from '../modules/nf-core/bcftools/index/main.nf'
 
 workflow RUN_GLIMPSE {
 
@@ -9,23 +11,32 @@ workflow RUN_GLIMPSE {
     vcf = channel.fromPath(params.vcf_in)
     vcf_samples = vcf.combine(SPLIT_SAMPLES.out.sample_lists.flatten())
 
-    SPLIT_VCFS(vcf_samples)
-    // // SPLIT_VCFS.out.split_vcfs.view()
+    vcf_samples_input = vcf_samples.map{
+	vcf, sample_list ->
+		[[id: vcf.getSimpleName() + "-" + sample_list.getName(), batch: sample_list.getName()], vcf, sample_list]
+}
+
+    SPLIT_VCFS(vcf_samples_input)
+    SPLIT_VCFS.out.split_vcfs.view()
 
     ref = channel.fromPath("${params.refdir}*.bin")
 
     phase_meta = [id: 'phase']
 
     phase_input = SPLIT_VCFS.out.split_vcfs.combine(ref).map{
-                                                                    vcf, index , ref_bin ->
-                                                                    [phase_meta, vcf, index, [], [], [], ref_bin, [], []]
+                                                                    meta, vcf, index , ref_bin ->
+                                                                    [meta, vcf, index, [], [], [], ref_bin, [], []]
                                                                 }
     phase_input2 = ['', params.fasta, params.fai]
 
     GLIMPSE2_PHASE(phase_input, phase_input2)
 
     GLIMPSE2_PHASE.out.versions.view()
 
+    INDEX_PHASE ( GLIMPSE2_PHASE.out.phased_variants )
+
+    // attempt for splitting samples in phase
+    ///
     // phase_input = vcf_samples.combine(ref).map {
     //                                                 vcf, samples, ref_bin ->
     //                                                 [phase_meta, vcf, [], samples, [], [], ref_bin, [], []]