Move to nf-core sbwf GLIMPSE1

CHANGELOG.md

@@ -8,6 +8,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ### `Added`
 
 - [#249](https://github.com/nf-core/phaseimpute/pull/249) - Add new `publish_all` hidden parameter to help during debugging. Add csv content checking in nf-test.
+- [#248](https://github.com/nf-core/phaseimpute/pull/248) - Add chromosomes concatenation to `BAM_GL_BCFTOOLS`
 
 ### `Changed`
 
@@ -20,6 +21,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - [#240](https://github.com/nf-core/phaseimpute/pull/240) - Move from local to nf-core sbwf for `GLIMPSE2` imputation. Update `usage.md`.
 - [#247](https://github.com/nf-core/phaseimpute/pull/243) - Move from local to nf-core sbwf for `BEAGLE5` imputation. Update `usage.md`.
 - [#249](https://github.com/nf-core/phaseimpute/pull/249) - Move `test_all`, `test_sim`, `test_panelprep`, `test_validate` to independant nf-test with md5sum assertion.
+- [#248](https://github.com/nf-core/phaseimpute/pull/248) - Move from local to nf-core sbwf for `GLIMPSE1` imputation. Set output channels to `vcf_index`. Update `usage.md`.
 
 ### `Fixed`
 

conf/steps/imputation_glimpse1.config

@@ -15,7 +15,12 @@ process {
 
     // Call the variants before imputation
     withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:GL_GLIMPSE1:.*' {
-        publishDir = [ enabled: false ]
+        publishDir = [
+            path: { "${params.outdir}/imputation/glimpse1/variant_calling" },
+            mode: params.publish_dir_mode,
+            enabled: params.publish_all,
+            saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+        ]
         tag = {"${meta.id} ${meta.chr}"}
     }
 
@@ -28,10 +33,13 @@ process {
 
     withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:GL_GLIMPSE1:BCFTOOLS_MERGE' {
         cache = "lenient"
-        ext.args = [
-            "--write-index=tbi",
-        ].join(' ')
-        ext.prefix = { "${meta.id}.merge" }
+        ext.args = "--write-index=tbi -Oz"
+        ext.prefix = { "${meta.id}_${meta.chr}.merge" }
+    }
+
+    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:GL_GLIMPSE1:VCF_CONCATENATE_BCFTOOLS:BCFTOOLS_CONCAT' {
+        ext.args = "--write-index=tbi -Oz"
+        ext.prefix = { "${meta.id}.concat" }
     }
 
     withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:GL_GLIMPSE1:BCFTOOLS_ANNOTATE' {
@@ -40,32 +48,33 @@ process {
     }
 
     // Impute the variants
-    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:VCF_IMPUTE_GLIMPSE1:.*' {
-        publishDir = [ enabled: false ]
-        tag = {"${meta.id} ${meta.chr}"}
+    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:VCF_IMPUTE_GLIMPSE:.*' {
+        publishDir = [
+            path: { "${params.outdir}/imputation/glimpse1/impute" },
+            mode: params.publish_dir_mode,
+            enabled: params.publish_all,
+            saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+        ]
+        tag = {"${meta.id} ${meta.regionout ?: meta.chr}"}
     }
 
-    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:VCF_IMPUTE_GLIMPSE1:GLIMPSE_PHASE' {
-        ext.args   = ["--impute-reference-only-variants"].join(' ')
-        ext.prefix = { "${meta.id}.batch${meta.batch}.${meta.chunk.replace(':','_')}.glimpse1" }
+    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:VCF_IMPUTE_GLIMPSE:GLIMPSE_PHASE' {
+        ext.args   = { "--impute-reference-only-variants --seed ${params.seed}" }
+        ext.prefix = { "${meta.id}.batch${meta.batch}.${meta.regionout ? meta.regionout.replace(':','_') : meta.chr}.glimpse1" }
         ext.suffix = "bcf"
-        publishDir = [ enabled: false ]
     }
 
-    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:VCF_IMPUTE_GLIMPSE1:BCFTOOLS_INDEX_1' {
+    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:VCF_IMPUTE_GLIMPSE:BCFTOOLS_INDEX_PHASE' {
         ext.args   = "--csi"
-        publishDir = [ enabled: false ]
     }
 
-    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:VCF_IMPUTE_GLIMPSE1:GLIMPSE_LIGATE' {
+    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:VCF_IMPUTE_GLIMPSE:GLIMPSE_LIGATE' {
         ext.args   = { "--seed ${params.seed}" }
         ext.prefix = { "${meta.id}.batch${meta.batch}.${meta.chr}.ligate.glimpse1" }
-        publishDir = [ enabled: false ]
     }
 
-    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:VCF_IMPUTE_GLIMPSE1:BCFTOOLS_INDEX_2' {
+    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:VCF_IMPUTE_GLIMPSE:BCFTOOLS_INDEX_LIGATE' {
         ext.args   = "--tbi"
-        publishDir = [ enabled: false ]
     }
 
     // Concatenate the imputed chunks

conf/steps/validation.config

@@ -28,19 +28,17 @@ process {
     }
 
     withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:GL_TRUTH:BCFTOOLS_MERGE' {
-        ext.args = [
-            "--write-index=tbi",
-        ].join(' ')
-        ext.prefix = { "${meta.id}" }
+        ext.args = "--write-index=tbi -Oz"
+        ext.prefix = { "${meta.id}_${meta.chr}" }
     }
 
-    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:GL_TRUTH:BCFTOOLS_ANNOTATE' {
-        ext.args   = ["--set-id '%CHROM:%POS:%REF:%ALT'", "-Oz", "--write-index=tbi"].join(' ')
-        ext.prefix = { "${meta.id}_${meta.chr}.annotate" }
+    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:GL_TRUTH:VCF_CONCATENATE_BCFTOOLS:BCFTOOLS_CONCAT' {
+        ext.args = "--write-index=tbi --output-type z --ligate"
+        ext.prefix = { "${meta.id}.batch${meta.batch}.concat" }
     }
 
-    // Concatenate the truth set
-    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:CONCAT_TRUTH:.*' {
+    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:GL_TRUTH:BCFTOOLS_ANNOTATE' {
+        ext.args   = ["--set-id '%CHROM:%POS:%REF:%ALT'", "-Oz", "--write-index=tbi"].join(' ')
         ext.prefix = { "${meta.id}.batch${meta.batch}.truth" }
         publishDir = [
             path: { "${params.outdir}/validation/concat" },
@@ -49,10 +47,6 @@ process {
         ]
     }
 
-    withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:CONCAT_TRUTH:BCFTOOLS_CONCAT' {
-        ext.args   = ["--ligate", "--output-type z", "--write-index=tbi"].join(' ')
-    }
-
     // Compute sample files for renaming
     withName: 'NFCORE_PHASEIMPUTE:PHASEIMPUTE:BCFTOOLS_QUERY_TRUTH' {
         tag = { "${meta.id} Batch ${meta.batch} ${meta.tools}" }

docs/usage.md

@@ -445,7 +445,7 @@ Optionnaly you can provide the following flags:
 
 |            | `--steps impute`(m) | `--input`(m) | `--genome` or `--fasta`(m) | `--panel`(m) | `--posfile`(m) | `--map`(o) | `--chunks`(o) |
 | ---------- | ------------------- | ------------ | -------------------------- | ------------ | -------------- | ---------- | ------------- |
-| `GLIMPSE1` | ✅                  | ✅ ¹         | ✅                         | ✅           | ✅ ³           | ❌⁶        | ✅            |
+| `GLIMPSE1` | ✅                  | ✅ ¹         | ✅                         | ✅           | ✅ ³           | ✅         | ✅            |
 | `GLIMPSE2` | ✅                  | ✅ ¹         | ✅                         | ✅           | ❌             | ✅         | ✅            |
 | `QUILT`    | ✅                  | ✅ ²         | ✅                         | ❌           | ✅ ⁴           | ❌⁶        | ✅            |
 | `STITCH`   | ✅                  | ✅ ²         | ✅                         | ❌           | ✅ ³           | ✅         | ✅            |
@@ -617,6 +617,13 @@ panel,chr,posfile
 
 The CSV file provided in `--panel` must be prepared with `--steps panelprep` and must contain four columns [panel, chr, vcf, index].
 
+You can optionally provide chunks to parallelize the imputation process using `--chunks`.
+If not provided the full region per chromosome will be used.
+See [Chunks section](#samplesheet-chunks) for more information.
+
+Genetic map can also be provided for better accuracy.
+See [Map section](#samplesheet-map) for more information.
+
 ### GLIMPSE2
 
 [GLIMPSE2](https://github.com/odelaneau/GLIMPSE) is a set of tools for phasing and imputation for low-coverage sequencing datasets. This is an example command to run this tool from the `--steps impute`:

modules.json

@@ -28,7 +28,8 @@
                             "bam_impute_stitch",
                             "bam_vcf_impute_glimpse2",
                             "modules",
-                            "vcf_impute_beagle5"
+                            "vcf_impute_beagle5",
+                            "vcf_impute_glimpse"
                         ]
                     },
                     "bcftools/merge": {
@@ -82,17 +83,17 @@
                     "glimpse/chunk": {
                         "branch": "master",
                         "git_sha": "41dfa3f7c0ffabb96a6a813fe321c6d1cc5b6e46",
-                        "installed_by": ["modules"]
+                        "installed_by": ["modules", "vcf_impute_glimpse"]
                     },
                     "glimpse/ligate": {
                         "branch": "master",
                         "git_sha": "41dfa3f7c0ffabb96a6a813fe321c6d1cc5b6e46",
-                        "installed_by": ["modules"]
+                        "installed_by": ["modules", "vcf_impute_glimpse"]
                     },
                     "glimpse/phase": {
                         "branch": "master",
                         "git_sha": "41dfa3f7c0ffabb96a6a813fe321c6d1cc5b6e46",
-                        "installed_by": ["modules"]
+                        "installed_by": ["modules", "vcf_impute_glimpse"]
                     },
                     "glimpse2/chunk": {
                         "branch": "master",
@@ -250,6 +251,11 @@
                         "git_sha": "4b406a74dc0449c0401ed87d5bfff4252fd277fd",
                         "installed_by": ["subworkflows"]
                     },
+                    "vcf_impute_glimpse": {
+                        "branch": "master",
+                        "git_sha": "6383d8fe58f9498eecd5aa303e71a4a932d1e9f6",
+                        "installed_by": ["subworkflows"]
+                    },
                     "vcf_impute_beagle5": {
                         "branch": "master",
                         "git_sha": "6383d8fe58f9498eecd5aa303e71a4a932d1e9f6",

subworkflows/local/bam_gl_bcftools/main.nf

@@ -1,6 +1,7 @@
-include { BCFTOOLS_MPILEUP          } from '../../../modules/nf-core/bcftools/mpileup'
-include { BCFTOOLS_MERGE            } from '../../../modules/nf-core/bcftools/merge'
-include { BCFTOOLS_ANNOTATE         } from '../../../modules/nf-core/bcftools/annotate'
+include { BCFTOOLS_MPILEUP                 } from '../../../modules/nf-core/bcftools/mpileup'
+include { BCFTOOLS_MERGE                   } from '../../../modules/nf-core/bcftools/merge'
+include { BCFTOOLS_ANNOTATE                } from '../../../modules/nf-core/bcftools/annotate'
+include { VCF_CONCATENATE_BCFTOOLS         } from '../../../subworkflows/local/vcf_concatenate_bcftools'
 
 workflow BAM_GL_BCFTOOLS {
 
@@ -34,7 +35,7 @@ workflow BAM_GL_BCFTOOLS {
         .map{ metaIPC, vcf, tbi -> [metaIPC.subMap("panel_id", "chr", "batch"), [metaIPC, vcf, tbi]] }
         .groupTuple(sort: { it1, it2 -> it1[0]["id"] <=> it2[0]["id"] }) // Sort by id
         .map{ metaPC, filestups -> [
-            metaPC + [id: "all", metas: filestups.collect{it -> it[0]}],
+            metaPC + [id: "all_samples", metas: filestups.collect{it -> it[0]}],
             filestups.collect{it -> it[1]},
             filestups.collect{it -> it[2]},
             filestups.collect{it -> it[1]}.size()
@@ -44,34 +45,42 @@ workflow BAM_GL_BCFTOOLS {
             more: it[3] > 1
         }
 
-    // Merge VCFs
+    // Merge VCFs all individuals
     BCFTOOLS_MERGE(
         ch_all_vcf.more.map{it -> [it[0], it[1], it[2], []] },
         ch_fasta
     )
     ch_versions = ch_versions.mix(BCFTOOLS_MERGE.out.versions.first())
 
     // Mix all vcfs
-    ch_to_annotate = ch_all_vcf.one
+    ch_to_concat = ch_all_vcf.one
         .map{it -> [it[0]["metas"][0], it[1][0], it[2][0]] }
         .mix(
             BCFTOOLS_MERGE.out.vcf
-                .join(BCFTOOLS_MERGE.out.tbi)
+                .join(BCFTOOLS_MERGE.out.tbi.mix(
+                    BCFTOOLS_MERGE.out.csi
+                ))
         )
 
+    // Merge all chromosomes
+    VCF_CONCATENATE_BCFTOOLS(ch_to_concat)
+    ch_versions = ch_versions.mix(VCF_CONCATENATE_BCFTOOLS.out.versions.first())
+
     // Annotate the variants
-    BCFTOOLS_ANNOTATE(ch_to_annotate
+    BCFTOOLS_ANNOTATE(VCF_CONCATENATE_BCFTOOLS.out.vcf_index
         .combine(channel.of([[], [], [], []]))
     )
     ch_versions = ch_versions.mix(BCFTOOLS_ANNOTATE.out.versions.first())
 
     // Output
     ch_output = BCFTOOLS_ANNOTATE.out.vcf
-        .join(BCFTOOLS_ANNOTATE.out.tbi)
-        .map{ metaIPC, vcf, tbi -> [metaIPC + [ variantcaller:'bcftools' ], vcf, tbi] }
+        .join(BCFTOOLS_ANNOTATE.out.tbi.mix(
+            BCFTOOLS_ANNOTATE.out.csi
+        ))
+        .map{ metaIPC, vcf, index -> [metaIPC + [ variantcaller:'bcftools' ], vcf, index] }
 
     emit:
-    vcf_tbi       = ch_output        // channel: [ [id, panel, chr], vcf, tbi ]
+    vcf_index     = ch_output        // channel: [ [id, panel], vcf, index ]
     versions      = ch_versions      // channel: [ versions.yml ]
     multiqc_files = ch_multiqc_files
 }

subworkflows/local/bam_gl_bcftools/tests/main.nf.test

@@ -18,7 +18,7 @@ nextflow_workflow {
     tag "bcftools/mpileup"
     tag "bcftools/annotate"
 
-    test("Compute genotype likelihood with merging") {
+    test("Compute genotype likelihood with merging and two chromosomes") {
         when {
             workflow {
                 """
@@ -34,10 +34,15 @@ nextflow_workflow {
                         file(params.pipelines_testdata_base_path + "hum_data/individuals/NA19401/NA19401.s.bam.bai", checkIfExist:true),
                     ],
                 ])
-                input[1] = channel.of([
-                        [panel_id: "1000GP", chr: "22"],
-                        file(params.pipelines_testdata_base_path + "hum_data/panel/chr22/1000GP.chr22.posfile", checkIfExist:true)
-                    ]).collect()
+                input[1] = channel.of(
+                [
+                    [panel_id: "1000GP", chr: "22"],
+                    file(params.pipelines_testdata_base_path + "hum_data/panel/chr22/1000GP.chr22.posfile", checkIfExist:true)
+                ],
+                [
+                    [panel_id: "1000GP", chr: "21"],
+                    file(params.pipelines_testdata_base_path + "hum_data/panel/chr21/1000GP.chr21.posfile", checkIfExist:true)
+                ])
                 input[2] = channel.of([
                     [id: "GRCh38"],
                     file(params.pipelines_testdata_base_path + "hum_data/reference_genome/GRCh38.s.fa.gz", checkIfExist:true),
@@ -52,12 +57,12 @@ nextflow_workflow {
                 { assert workflow.success },
                 { assert snapshot(
                     workflow.out.versions,
-                    workflow.out.vcf_tbi.collect{[
+                    workflow.out.vcf_index.collect{[
                         it[0],
                         path(it[1]).getFileName().toString(),
                         path(it[2]).getFileName().toString()
                     ] },
-                    workflow.out.vcf_tbi.collect{
+                    workflow.out.vcf_index.collect{
                         path(it[1]).vcf.summary
                     },
                     workflow.out.versions.collect{ path(it).yaml }
@@ -95,12 +100,12 @@ nextflow_workflow {
                 { assert workflow.success },
                 { assert snapshot(
                     workflow.out.versions,
-                    workflow.out.vcf_tbi.collect{[
+                    workflow.out.vcf_index.collect{[
                         it[0],
                         path(it[1]).getFileName().toString(),
                         path(it[2]).getFileName().toString()
                     ] },
-                    workflow.out.vcf_tbi.collect{
+                    workflow.out.vcf_index.collect{
                         path(it[1]).vcf.summary
                     },
                     workflow.out.versions.collect{ path(it).yaml }