Long-read GENS PON

CHANGELOG.md

@@ -38,6 +38,7 @@ Initial release of nf-core/createpanelrefs, created with the [nf-core](https://n
 - [#54](https://github.com/nf-core/createpanelrefs/pull/54) - Update nft-utils to 0.0.4
 - [#55](https://github.com/nf-core/createpanelrefs/pull/55) - Prepare relase 1.0.0
 - [#63](https://github.com/nf-core/createpanelrefs/pull/63) - Template update for nf-core/tools v3.5.0dev
+- [#66](https://github.com/nf-core/createpanelrefs/pull/66) - Update `GENS` to allow for creating a long-read PON
 
 ### `Fixed`
 
@@ -56,6 +57,7 @@ Initial release of nf-core/createpanelrefs, created with the [nf-core](https://n
 | gatk4      |             | 4.6.2.0     |
 | gawk       |             | 5.3.0       |
 | htslib     |             | 1.22.1      |
+| mosdepth   |             | 0.3.11      |
 | multiqc    |             | 1.32        |
 | samtools   |             | 1.22.1      |
 

conf/modules/gens_pon.config

@@ -17,22 +17,70 @@ process {
         ]
     }
 
+    withName: '.*GENS_PON:CAT_CAT' {
+        ext.prefix = { "${meta.id}_concat" }
+    }
+
     withName: '.*GENS_PON:GATK4_COLLECTREADCOUNTS' {
-        ext.args = {"--format ${params.gens_readcount_format} --imr OVERLAPPING_ONLY"}
+        ext.args   = { "--format ${params.gens_readcount_format} --imr OVERLAPPING_ONLY" }
         publishDir = [
-                mode: params.publish_dir_mode,
-                path: { "${params.outdir}/gens_pon/readcounts" },
-                saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+            mode: params.publish_dir_mode,
+            path: { "${params.outdir}/gens_pon/readcounts" },
+            saveAs: { filename -> filename.equals('versions.yml') ? null : filename },
+        ]
+    }
+
+    withName: '.*GENS_PON:INTERVAL_LIST_TO_BED' {
+        ext.prefix = { "gens_coverage_bins" }
+        ext.suffix = "bed"
+        ext.args2 = '\'BEGIN { FS=OFS="\t" } $2 < $3 { print $1, $2, $3 }\''
+        publishDir = [
+            mode: params.publish_dir_mode,
+            path: { "${params.outdir}/references/intervals/gens_pon/" },
+            saveAs: { filename -> filename.equals('versions.yml') ? null : filename },
         ]
     }
 
     withName: '.*GENS_PON:GATK4_CREATEREADCOUNTPANELOFNORMALS' {
-        ext.args   = { ["--minimum-interval-median-percentile ${params.gens_min_interval_median_percentile}",
-                        "--maximum-chunk-size ${params.gens_maximum_chunk_size}"].join(" ")}
+        ext.args   = {
+            [
+                "--minimum-interval-median-percentile ${params.gens_min_interval_median_percentile}",
+                "--maximum-chunk-size ${params.gens_maximum_chunk_size}",
+            ].join(" ")
+        }
         publishDir = [
-                mode: params.publish_dir_mode,
-                path: { "${params.outdir}/gens_pon/createreadcountpanelofnormals" },
-                saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+            mode: params.publish_dir_mode,
+            path: { "${params.outdir}/gens_pon/createreadcountpanelofnormals" },
+            saveAs: { filename -> filename.equals('versions.yml') ? null : filename },
         ]
     }
+
+    withName: '.*GENS_PON:MOSDEPTH' {
+        ext.args = {
+            [
+                '--no-per-base',
+                '--fast-mode',
+            ].join(' ')
+        }
+    }
+
+    withName: '.*GENS_PON:MOSDEPTH_GATK_FORMAT' {
+        ext.prefix = { "${meta.id}_gatk_formatted" }
+        ext.suffix = "gatk_format.tsv"
+        ext.args   = { "-v sample=${meta.id}" }
+        ext.args2  = '\'BEGIN { OFS="\\t" } { $4 = int($4 + 0.5); $2++; print $1, $2, $3, $4 }\''
+    }
+
+    withName: '.*GENS_PON:MOSDEPTH_GATK_HEADER' {
+        // This requires meta.id to be the same as the sample name in the BAM/CRAM header
+        ext.args   = { "-v sample=${meta.id}" }
+        ext.args2  = '\'{ print } END { print "@RG\\tID:GATKCopyNumber\\tSM:" sample; print "CONTIG\\tSTART\\tEND\\tCOUNT" }\''
+        ext.prefix = { "${meta.id}" }
+        ext.suffix = 'mosdepth_gatk_header.tsv'
+    }
+
+    withName: '.*GENS_PON:SAMTOOLS_VIEW' {
+        ext.args   = '-H --output-fmt sam'
+        ext.prefix = { "${meta.id}.mosdepth_gatk_header" }
+    }
 }

docs/output.md

@@ -88,11 +88,14 @@ The reference file contains coverage information normalized across the cohort an
     - `{pon_name}.hdf5`: Final panel of normals file in HDF5 format.
   - `references/intervals/gens_pon/`
     - `*.interval_list`: Interval list file used for read count collection.
+    - `*.bed`: BED versions of interval list file used for read count collection for long-reads.
 
 </details>
 
 [GENS](https://github.com/Clinical-Genomics-Lund/gens) creates a panel of normals for read-count denoising to improve somatic variant detection. The workflow: (1) indexes BAM/CRAM files if needed, (2) collects read counts at specified intervals using GATK's CollectReadCounts, and (3) creates a panel of normals using GATK's CreateReadCountPanelOfNormals. This panel can be used with GENS for somatic variant calling to reduce technical noise and improve variant detection sensitivity.
 
+When `gens_analysis_type` is set to 'lrs', a modified version of the workflow above is run where coverage calculated by mosdepth is used instead of read counts.
+
 ### MultiQC
 
 <details markdown="1">

docs/usage.md

@@ -131,13 +131,17 @@ If you are running the pipeline to generate references for the GENS workflow, yo
 | ------------- | --------------------- |
 | fasta/genomes | fai                   |
 |               | dict                  |
+|               | gens_analysis_type    |
 |               | gens_bin_length       |
 |               | gens_pon_name         |
 |               | gens_readcount_format |
 |               | gens_interval_list    |
 
 The GENS workflow creates a panel of normals for read-count denoising from normal samples. This panel can be used with GENS for somatic variant calling to reduce technical noise and improve variant detection. For more information, see the [GENS documentation](https://github.com/Clinical-Genomics-Lund/gens).
 
+> [!NOTE]
+> If `--gens_analysis_type` is set to 'lrs', this reqires the sample ID set in the sample sheet to be equal to the `SM` tag in the BAM-file.
+
 ### germlinecnvcaller
 
 If you are running the pipeline to generate references for the GATK's germlinecnvcalling workflow, you should ensure that you have provided all the mandatory options specified in the table below.

main.nf

@@ -137,6 +137,7 @@ workflow {
         PIPELINE_INITIALISATION.out.samplesheet,
         params.tools ?: "no_tools",
         params.gcnv_model_name,
+        params.gens_analysis_type,
         params.gens_pon_name,
         params.mutect2_pon_name,
         fasta,
@@ -232,6 +233,7 @@ workflow NFCORE_CREATEPANELREFS {
     samplesheet // channel: samplesheet read in from --input
     tools // string: comma separated list of tools to run
     gcnv_model_name // string: name of gcnv model
+    gens_analysis_type // string: type of analysis for gens pon ('lrs' or 'srs')
     gens_pon_name // string: name of gens pon
     mutect2_pon_name // string: name of mutect2 pon
     fasta // channel: [meta, fasta]
@@ -250,7 +252,7 @@ workflow NFCORE_CREATEPANELREFS {
 
     main:
     // WORKFLOW: Run pipeline
-    CREATEPANELREFS(samplesheet, tools, gcnv_model_name, gens_pon_name, mutect2_pon_name, fasta, dict, fai, cnvkit_targets, gcnv_exclude_bed, gcnv_exclude_interval_list, gcnv_mappable_regions, gcnv_ploidy_priors, gcnv_segmental_duplications, gcnv_target_bed, gcnv_target_interval_list, gens_interval_list, mutect2_target_bed)
+    CREATEPANELREFS(samplesheet, tools, gcnv_model_name, gens_analysis_type, gens_pon_name, mutect2_pon_name, fasta, dict, fai, cnvkit_targets, gcnv_exclude_bed, gcnv_exclude_interval_list, gcnv_mappable_regions, gcnv_ploidy_priors, gcnv_segmental_duplications, gcnv_target_bed, gcnv_target_interval_list, gens_interval_list, mutect2_target_bed)
 
     emit:
     versions = CREATEPANELREFS.out.versions // channel: versions.yml

modules.json

@@ -5,6 +5,11 @@
         "https://github.com/nf-core/modules.git": {
             "modules": {
                 "nf-core": {
+                    "cat/cat": {
+                        "branch": "master",
+                        "git_sha": "69614d4579a6bd9b8a2ecffb35959809d9c36559",
+                        "installed_by": ["modules"]
+                    },
                     "cnvkit/batch": {
                         "branch": "master",
                         "git_sha": "09223d6de1dab602242c4c57ab2a4599d460e528",
@@ -85,6 +90,11 @@
                         "git_sha": "41dfa3f7c0ffabb96a6a813fe321c6d1cc5b6e46",
                         "installed_by": ["modules"]
                     },
+                    "mosdepth": {
+                        "branch": "master",
+                        "git_sha": "6832b69ef7f98c54876d6436360b6b945370c615",
+                        "installed_by": ["modules"]
+                    },
                     "multiqc": {
                         "branch": "master",
                         "git_sha": "af27af1be706e6a2bb8fe454175b0cdf77f47b49",