Deepvariant

.gitignore

@@ -9,3 +9,4 @@ testing*
 null/
 .nf-test
 .nf-test.log
+out/

conf/base.config

@@ -26,6 +26,26 @@ process {
     //        adding in your local modules too.
     // TODO nf-core: Customise requirements for specific processes.
     // See https://www.nextflow.io/docs/latest/config.html#config-process-selectors
+    withLabel:process_gpu_high {
+        ext.use_gpu = { params.use_gpu as boolean }
+        accelerator = { (params.use_gpu as boolean) ? 1 : null }
+        cpus   = { 8      * task.attempt }
+        memory = { 48.GB  * task.attempt }
+        time   = { 8.h   * task.attempt }
+    }
+    withLabel:process_gpu_very_high {
+        ext.use_gpu = { params.use_gpu as boolean }
+        accelerator = { (params.use_gpu as boolean) ? 1 : null }
+        cpus   = { 16     * task.attempt }
+        memory = { 96.GB  * task.attempt }
+        time   = { 10.h   * task.attempt }
+    }
+    withLabel:process_gpu_very_high_memory {
+        ext.use_gpu = { params.use_gpu as boolean }
+        accelerator = { (params.use_gpu as boolean) ? 1 : null }
+        cpus   = { 16     * task.attempt }
+        memory = { 128.GB * task.attempt }
+    }
     withLabel:process_single {
         cpus   = { 1                   }
         memory = { 6.GB   * task.attempt }
@@ -64,8 +84,4 @@ process {
         errorStrategy = 'retry'
         maxRetries    = 2
     }
-    withLabel: process_gpu {
-        ext.use_gpu = { workflow.profile.contains('gpu') }
-        accelerator = { workflow.profile.contains('gpu') ? 1 : null }
-    }
 }

conf/igenomes.config

@@ -43,6 +43,7 @@ params {
             dbsnp           = "${params.igenomes_base}/Homo_sapiens/ClairSTO/CHM13/Annotation/ClairSTO-pon/final_dbsnp.vcf.gz"
             onekgenomes     = "${params.igenomes_base}/Homo_sapiens/ClairSTO/CHM13/Annotation/ClairSTO-pon/final_1kgenomes.vcf.gz"
             colors          = "${params.igenomes_base}/Homo_sapiens/ClairSTO/CHM13/Annotation/ClairSTO-pon/final_colors.vcf.gz"
+            asap            = "${params.igenomes_base}/Homo_sapiens/ClairSTO/CHM13/Annotation/ClairSTO-pon/WGS_CHM13_ASAP.vcf.gz"
         }
         'GRCm38' {
             fasta       = "${params.igenomes_base}/Mus_musculus/Ensembl/GRCm38/Sequence/WholeGenomeFasta/genome.fa"

conf/modules.config

@@ -116,6 +116,121 @@ process {
         ]
     }
 
+    withName: '.*:BCFTOOLS_NORM' {
+        ext.prefix = { "${meta.id}.${meta.caller}_norm" }
+        ext.args = {
+            "-Oz -W=tbi"
+        }
+        publishDir = [
+            enabled: false
+        ]
+    }
+
+    withName: '.*:BCFTOOLS_ISEC' {
+        ext.prefix = { "${meta.id}_isec" }
+        ext.args ={
+            "-Oz"
+        }
+        publishDir = [
+            enabled: false
+        ]
+    }
+    withName: '.*:BCFTOOLS_ANNOTATE' {
+        ext.prefix = { "${meta.id}.${meta.caller}" }
+        ext.args = {
+            '''-h <(echo '##INFO=<ID=CALLER,Number=1,Type=String,Description="Variant caller">') \
+                -c CHROM,POS,REF,ALT,INFO/CALLER \
+                -Oz \
+                -W=tbi'''
+        }
+        publishDir = [
+            enabled: false
+        ]
+    }
+    withName: '.*:BCFTOOLS_QUERY' {
+        ext.args = {
+            "-f '%CHROM\t%POS\t%REF\t%ALT\t${meta.caller}\n'"
+        }
+        publishDir = [
+            enabled: false
+        ]
+    }
+
+    withName: '.*DEEPVARIANT_MAKEEXAMPLES' {
+        ext.args = {
+            meta.platform == 'pb'
+                ? '--channel_list "BASE_CHANNELS,haplotype,base_6ma" --alt_aligned_pileup "diff_channels" --pileup_image_width "147"'
+                : '--channel_list "BASE_CHANNELS,haplotype" --alt_aligned_pileup "diff_channels" --pileup_image_width "99"'
+        }
+        publishDir = [
+            enabled: false
+        ]
+    }
+
+    withName: '.*DEEPVARIANT_POSTPROCESSVARIANTS' {
+        publishDir = [
+            path: { "${params.outdir}/${meta.id}/variants/deepvariant" },
+            mode: params.publish_dir_mode,
+            saveAs: { filename ->
+                (filename.equals('versions.yml') || filename.endsWith('.g.vcf.gz') || filename.endsWith('.g.vcf.gz.tbi')) ? null : filename
+            }
+        ]
+
+    }
+
+    withName: '.*DEEPVARIANT_CALLVARIANTS' {
+        ext.args = {
+            meta.platform == 'pb' ? ("--checkpoint '/opt/models/pacbio' ") : ("--checkpoint '/opt/models/ont_r104'")
+        }
+        publishDir = [
+            enabled : false
+        ]
+    }
+
+    withName: '.*DEEPSOMATIC_MAKEEXAMPLES' {
+        ext.args = {
+            meta.platform == 'pb'
+                ? meta.paired_data
+                    ? '--channel_list "BASE_CHANNELS,haplotype" --alt_aligned_pileup "diff_channels" --pileup_image_width "99"'
+                    : '--channel_list "BASE_CHANNELS,haplotype,allele_frequency" --alt_aligned_pileup "diff_channels" --pileup_image_width "99" --population_vcfs "/opt/models/deepsomatic/pons/AF_pacbio_PON_CoLoRSdb.GRCh38.AF0.05.vcf.gz","AF_ilmn_PON_DeepVariant.GRCh38.AF0.05.vcf.gz","PON_dbsnp138_gnomad_ILMN1000g_pon.vcf.gz","PON_dbsnp138_gnomad_PB1000g_pon.vcf.gz"'
+                : meta.paired_data
+                    ? '--channel_list "BASE_CHANNELS,haplotype" --alt_aligned_pileup "diff_channels" --pileup_image_width "99"'
+                    : '--channel_list "BASE_CHANNELS,haplotype,allele_frequency" --alt_aligned_pileup "diff_channels" --pileup_image_width "99" --population_vcfs "/opt/models/deepsomatic/pons/AF_pacbio_PON_CoLoRSdb.GRCh38.AF0.05.vcf.gz","AF_ilmn_PON_DeepVariant.GRCh38.AF0.05.vcf.gz","PON_dbsnp138_gnomad_ILMN1000g_pon.vcf.gz","PON_dbsnp138_gnomad_PB1000g_pon.vcf.gz"'
+        }
+        publishDir = [
+            enabled: false
+        ]
+    }
+
+    withName: '.*DEEPSOMATIC_POSTPROCESSVARIANTS' {
+        ext.args = {
+            '--process_somatic=true'
+        }
+        publishDir = [
+            path: { "${params.outdir}/${meta.id}/variants/deepsomatic" },
+            mode: params.publish_dir_mode,
+            saveAs: { filename ->
+                (filename.equals('versions.yml') || filename.endsWith('.g.vcf.gz') || filename.endsWith('.g.vcf.gz.tbi')) ? null : filename
+            }
+        ]
+    }
+
+    withName: '.*DEEPSOMATIC_CALLVARIANTS' {
+        ext.args = {
+            meta.platform == 'pb'
+                ? (meta.paired_data
+                    ? "--checkpoint '/opt/models/deepsomatic/pacbio'"
+                    : "--checkpoint '/opt/models/deepsomatic/pacbio_tumor_only'" )
+                : (meta.paired_data
+                    ? "--checkpoint '/opt/models/deepsomatic/ont'"
+                    : "--checkpoint '/opt/models/deepsomatic/ont_tumor_only'")
+        }
+        publishDir = [
+            enabled : false
+        ]
+    }
+
+
     withName: '.*:UNZIP_.*' {
         publishDir = [
             enabled: false
@@ -161,7 +276,7 @@ process {
     //
 
     withName: '.*:MINIMAP2_ALIGN' {
-        ext.prefix = { "${meta.id}_mapped" }
+        ext.prefix = { "${meta.id}_${meta.type}_mapped" }
         ext.args = {
             [
                 meta.platform == 'pb' ? ( params.minimap2_pb_model ? "-ax $params.minimap2_pb_model" : "-ax map-hifi" ) :
@@ -218,7 +333,7 @@ process {
     // Phasing processes
     //
 
-    withName: '.*:LONGPHASE_PHASE' {
+    withName: '.*:LONGPHASE_PHASE_GERMLINE' {
         ext.prefix = { "germline_smallvariants" }
         ext.args   = {
                 [
@@ -232,15 +347,33 @@ process {
             saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
         ]
     }
-
-    withName: '.*:LONGPHASE_HAPLOTAG' {
+    withName: '.*:LONGPHASE_PHASE_SOMATIC' {
+        ext.prefix = { "somatic_smallvariants" }
+        ext.args   = {
+                [
+                    meta.platform == 'pb' ? '--pb' : '--ont',
+                    "--indels",
+                ].join(' ').trim()
+            }
+        publishDir = [
+            path: { "${params.outdir}/${meta.id}/variants/phased" },
+            mode: params.publish_dir_mode,
+            saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+        ]
+    }
+    withName: '.*:LONGPHASE_HAPLOTAG*' {
         ext.prefix = { "${meta.id}_${meta.type}" }
         publishDir = [
             path: { "${params.outdir}/${meta.id}/bamfiles" },
             mode: params.publish_dir_mode,
             saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
         ]
     }
+    withName: '.*:LONGPHASE_MODCALL.*' {
+        publishDir = [
+            enabled: false
+        ]
+    }
 
     //
     // Structural variant calling processes
@@ -260,14 +393,35 @@ process {
     // Small variant calling processes
     //
     withName: '.*:BCFTOOLS_CONCAT' {
-        ext.args = '--output-type z -a'
+        ext.args = {'-Oz -a -W=tbi'}
         publishDir = [
             enabled: false
         ]
     }
     withName: '.*:BCFTOOLS_SORT' {
         ext.prefix = { "${meta.id}_sorted" }
-        ext.arge = '--output-type z'
+        ext.args = {'-Oz -W=tbi'}
+        publishDir = [
+            enabled: false
+        ]
+    }
+    withName: '.*:GERMLINE_CONSENSUS:BCFTOOLS_SORT' {
+        ext.prefix = { "${meta.id}_germline_sorted" }
+        ext.args = {'-Oz -W=tbi'}
+        publishDir = [
+            enabled: false
+        ]
+    }
+    withName: '.*:SOMATIC_CONSENSUS:BCFTOOLS_SORT' {
+        ext.prefix = { "${meta.id}_somatic_sorted" }
+        ext.args = {'-Oz -W=tbi'}
+        publishDir = [
+            enabled: false
+        ]
+    }
+    withName: '.*:PAIRED_SMALLVAR_SOMATIC:BCFTOOLS_SORT' {
+        ext.prefix = { "${meta.id}_somatic_sorted" }
+        ext.args = {'-Oz -W=tbi'}
         publishDir = [
             enabled: false
         ]

conf/test.config

@@ -18,6 +18,29 @@ process {
             time: '1.h'
         ]
     }
+
+    withName: '.*DEEPVARIANT_MAKEEXAMPLES' {
+        ext.args = {
+           "--regions 'chr19'"
+        }
+    }
+
+    withName: '.*DEEPVARIANT_POSTPROCESSVARIANTS' {
+        ext.args = {
+            "--regions 'chr19'"
+        }
+    }
+    withName: '.*DEEPSOMATIC_MAKEEXAMPLES' {
+       ext.args = {
+            "--regions 'chr19'"
+       }
+    }
+    withName: '.*DEEPSOMATIC_POSTPROCESSVARIANTS' {
+       ext.args = {
+            "--regions 'chr19'"
+       }
+    }
+
 }
 
 params {
@@ -29,7 +52,7 @@ params {
     fasta    = "https://raw.githubusercontent.com/IntGenomicsLab/test-datasets/main/references/GRCh38_chr19.fasta.gz"
 
     // Additional params
-    genome             = "GRCh38"
+    genome             = "CHM13"
     vep_genome         = "WBcel235"
     vep_species        = "caenorhabditis_elegans"
     skip_wakhan        = true

modules.json

@@ -11,16 +11,39 @@
                         "installed_by": ["modules"],
                         "patch": "modules/nf-core/ascat/ascat.diff"
                     },
+                    "bcftools/annotate": {
+                        "branch": "master",
+                        "git_sha": "3d9c2f4beaa4f62b3f006928fd9095a496d1e5a8",
+                        "installed_by": ["modules"]
+                    },
                     "bcftools/concat": {
                         "branch": "master",
                         "git_sha": "6383d8fe58f9498eecd5aa303e71a4a932d1e9f6",
                         "installed_by": ["modules", "vcf_gather_bcftools"]
                     },
+                    "bcftools/isec": {
+                        "branch": "master",
+                        "git_sha": "3b2c3559699a7bca6a7c2b220695a072e030e17d",
+                        "installed_by": ["modules"],
+                        "patch": "modules/nf-core/bcftools/isec/bcftools-isec.diff"
+                    },
                     "bcftools/merge": {
                         "branch": "master",
                         "git_sha": "3d9c2f4beaa4f62b3f006928fd9095a496d1e5a8",
+                        "installed_by": ["modules"],
+                        "patch": "modules/nf-core/bcftools/merge/bcftools-merge.diff"
+                    },
+                    "bcftools/norm": {
+                        "branch": "master",
+                        "git_sha": "6383d8fe58f9498eecd5aa303e71a4a932d1e9f6",
                         "installed_by": ["modules"]
                     },
+                    "bcftools/query": {
+                        "branch": "master",
+                        "git_sha": "6383d8fe58f9498eecd5aa303e71a4a932d1e9f6",
+                        "installed_by": ["modules"],
+                        "patch": "modules/nf-core/bcftools/query/bcftools-query.diff"
+                    },
                     "bcftools/sort": {
                         "branch": "master",
                         "git_sha": "6383d8fe58f9498eecd5aa303e71a4a932d1e9f6",
@@ -30,17 +53,20 @@
                     "deepvariant/callvariants": {
                         "branch": "master",
                         "git_sha": "f2b138ee1d91f67d31c187317d7e83e429bf0309",
-                        "installed_by": ["deepvariant"]
+                        "installed_by": ["deepvariant"],
+                        "patch": "modules/nf-core/deepvariant/callvariants/deepvariant-callvariants.diff"
                     },
                     "deepvariant/makeexamples": {
                         "branch": "master",
                         "git_sha": "f2b138ee1d91f67d31c187317d7e83e429bf0309",
-                        "installed_by": ["deepvariant"]
+                        "installed_by": ["deepvariant"],
+                        "patch": "modules/nf-core/deepvariant/makeexamples/deepvariant-makeexamples.diff"
                     },
                     "deepvariant/postprocessvariants": {
                         "branch": "master",
                         "git_sha": "f2b138ee1d91f67d31c187317d7e83e429bf0309",
-                        "installed_by": ["deepvariant"]
+                        "installed_by": ["deepvariant"],
+                        "patch": "modules/nf-core/deepvariant/postprocessvariants/deepvariant-postprocessvariants.diff"
                     },
                     "ensemblvep/download": {
                         "branch": "master",
@@ -55,12 +81,13 @@
                     },
                     "longphase/haplotag": {
                         "branch": "master",
-                        "git_sha": "e753770db613ce014b3c4bc94f6cba443427b726",
-                        "installed_by": ["modules"]
+                        "git_sha": "b8d30a43f33aee3148b0e9e9f00587984a4ac195",
+                        "installed_by": ["modules"],
+                        "patch": "modules/nf-core/longphase/haplotag/longphase-haplotag.diff"
                     },
                     "longphase/phase": {
                         "branch": "master",
-                        "git_sha": "47983538e45e539f783ed8ab0d1c96d39df2af8f",
+                        "git_sha": "b8d30a43f33aee3148b0e9e9f00587984a4ac195",
                         "installed_by": ["modules"],
                         "patch": "modules/nf-core/longphase/phase/longphase-phase.diff"
                     },
@@ -166,6 +193,12 @@
                         "git_sha": "7ac6cbe7c17c2dad685da7f70496c8f48ea48687",
                         "installed_by": ["subworkflows"]
                     },
+                    "deepvariant": {
+                        "branch": "master",
+                        "git_sha": "f2b138ee1d91f67d31c187317d7e83e429bf0309",
+                        "installed_by": ["subworkflows"],
+                        "patch": "subworkflows/nf-core/deepvariant/deepvariant.diff"
+                    },
                     "utils_nextflow_pipeline": {
                         "branch": "master",
                         "git_sha": "05954dab2ff481bcb999f24455da29a5828af08d",