Add GRIDSS preprocessing (#175)

* add gridss nf module

* add preprocess gridss BAM process and update I/O/Script section

* add gridss version and  docker

* add preprocess_BAM_GRIDSS

* add gridss_reference_diir param to template.config

* input preprocess channels; set variables; publish formatted filenames

* add gridss2 channels, ref and process call

* add gridss2 to algorithhms in template

* update schema YAML

* add gridss ref dir to schema

* Update CHANGELOG.md

* update metadata.yaml

* add jvmheap memory arg to gridss

* 4 CPUs and 10GB memory for gridss preprocessing

* change gridss ref dir to ref FASTA

* change gridss ref dir to ref FASTA

* reconfigure gridss ref files variable

* fix gridss ref file path

* store intermediate file files in their process and store sample files in sample dirs

* save logs tintask index dir

* add F32, F72 and M64 configs

* update memory in F32, F72 and M64

---------

Co-authored-by: Mootor <mmootor@ip-0A125937.rhxrlfvjyzbupc03cc22jkch3c.xx.internal.cloudapp.net>

CHANGELOG.md

@@ -9,15 +9,14 @@ This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.htm
 
 ## [Unreleased]
 ### Added
+- Add GRIDSS2 preprocessing
 - Add supported Nextflow version to `README.md`
+- Add PlantUML diagram
 
 ### Changed
 - Update PlantUML action to `v1.0.1`
 - Update memory allocations in `M64.config`
 
-### Added
-- Add PlantUML diagram
-
 ---
 
 ## [6.1.0] - 2024-03-12

config/F16.config

@@ -4,6 +4,17 @@ process {
  memory = 1.GB
  }
 
+ withName: preprocess_BAM_GRIDSS {
+ cpus = 4
+ memory = 10.GB
+ retry_strategy {
+ memory {
+ strategy = 'exponential'
+ operand = 2
+ }
+ }
+ }
+
  withName: call_sSV_Delly {
  cpus = 1
  memory = 16.GB
@@ -36,7 +47,7 @@ process {
  }
  }
  }
- 
+
  withName: call_sSV_Manta {
  cpus = 1
  memory = 16.GB

config/F32.config

@@ -4,6 +4,17 @@ process {
  memory = 1.GB
  }
 
+ withName: preprocess_BAM_GRIDSS {
+ cpus = 4
+ memory = 10.GB
+ retry_strategy {
+ memory {
+ strategy = 'exponential'
+ operand = 2
+ }
+ }
+ }
+
  withName: call_sSV_Delly {
  cpus = 1
  memory = 30.GB
@@ -36,7 +47,7 @@ process {
  }
  }
  }
- 
+
  withName: call_sSV_Manta {
  cpus = 1
  memory = 30.GB

config/F72.config

@@ -4,6 +4,17 @@ process {
  memory = 1.GB
  }
 
+ withName: preprocess_BAM_GRIDSS {
+ cpus = 4
+ memory = 15.GB
+ retry_strategy {
+ memory {
+ strategy = 'exponential'
+ operand = 2
+ }
+ }
+ }
+
  withName: call_sSV_Delly {
  cpus = 1
  memory = 30.GB
@@ -36,7 +47,7 @@ process {
  }
  }
  }
- 
+
  withName: call_sSV_Manta {
  cpus = 1
  memory = 30.GB

config/M64.config

@@ -4,6 +4,17 @@ process {
  memory = 1.GB
  }
 
+ withName: preprocess_BAM_GRIDSS {
+ cpus = 4
+ memory = 100.GB
+ retry_strategy {
+ memory {
+ strategy = 'exponential'
+ operand = 2
+ }
+ }
+ }
+
  withName: call_sSV_Delly {
  cpus = 1
  memory = 120.GB
@@ -36,7 +47,7 @@ process {
  }
  }
  }
- 
+
  withName: call_sSV_Manta {
  cpus = 1
  memory = 30.GB

config/default.config

@@ -27,12 +27,14 @@ params {
  // Pipeline tool versions
  delly_version = '1.2.6'
  manta_version = '1.6.0'
+ gridss_version = '2.13.2'
  bcftools_version = '1.15.1'
  pipeval_version = '4.0.0-rc.2'
 
  // Docker tool versions
  docker_image_delly = "${-> params.docker_container_registry}/delly:${params.delly_version}"
  docker_image_manta = "${-> params.docker_container_registry}/manta:${params.manta_version}"
+ docker_image_gridss = "${-> params.docker_container_registry}/gridss:${params.gridss_version}"
  docker_image_bcftools = "${-> params.docker_container_registry}/bcftools:${params.bcftools_version}"
  docker_image_validate = "${-> params.docker_container_registry}/pipeval:${params.pipeval_version}"
  }

config/schema.yaml

@@ -8,6 +8,11 @@ reference_fasta:
  mode: 'r'
  required: true
  help: 'Absolute path to a reference FASTA file'
+gridss_reference_fasta:
+ type: 'Path'
+ mode: 'r'
+ required: true
+ help: 'Absolute path to GRIDSS2 reference FASTA file'
 exclusion_file:
  type: 'Path'
  mode: 'r'
@@ -20,9 +25,11 @@ algorithm:
  default:
  - delly
  - manta
+ - gridss2
  choices:
  - delly
  - manta
+ - gridss2
 output_dir:
  type: 'Path'
  mode: 'w'

config/template.config

@@ -12,13 +12,14 @@ params {
  blcds_registered_dataset = false
 
  reference_fasta = "/hot/ref/reference/GRCh38-BI-20160721/Homo_sapiens_assembly38.fasta"
+ gridss_reference_fasta = "/hot/ref/tool-specific-input/GRIDSS-2.13.2/GRCh38-BI-20160721/Homo_sapiens_assembly38.fasta"
 
  exclusion_file = "/hot/ref/tool-specific-input/Delly/hg38/human.hg38.excl.tsv"
 
  output_dir = "where/to/save/outputs/"
 
  // select the tool(s) to run
- algorithm = [] // algorithm = ['delly', 'manta']
+ algorithm = [] // algorithm = ['delly', 'manta', 'gridss2']
 
  save_intermediate_files = false
 

main.nf

@@ -34,6 +34,7 @@ Current Configuration:
  DELLY: ${params.delly_version}
  BCFtools: ${params.bcftools_version}
  Manta: ${params.manta_version}
+ GRIDSS2: ${params.gridss_version}
  PipeVal: ${params.pipeval_version}
 
 ------------------------------------
@@ -54,6 +55,9 @@ include { call_sSV_Delly; filter_sSV_Delly } from './module/delly' addParams(
 include { call_sSV_Manta } from './module/manta' addParams(
  workflow_output_dir: "${params.output_dir_base}/Manta-${params.manta_version}"
  )
+include { preprocess_BAM_GRIDSS } from './module/gridss' addParams(
+ workflow_output_dir: "${params.output_dir_base}/GRIDSS-${params.gridss_version}"
+ )
 include { generate_sha512 as generate_sha512_BCFtools } from './module/sha512' addParams(
  workflow_output_dir: "${params.output_dir_base}/DELLY-${params.delly_version}"
  )
@@ -74,10 +78,15 @@ Channel.from(params.samples_to_process)
  .map{ sample -> ['index': indexFile(sample.path)] + sample }
  .set{ input_ch_samples_with_index }
 
+Channel.from(params.samples_to_process)
+ .map{ sample -> [sample.id, sample.path, indexFile(sample.path)] }
+ .set{ gridss_ch }
+
 input_ch_samples_with_index
  .map{ sample -> [sample.path, sample.index] }
  .flatten()
  .set{ input_validation }
+
 if (params.verbose){
  input_validation.view()
  }
@@ -96,6 +105,9 @@ if (params.verbose){
 
 reference_fasta_index = "${params.reference_fasta}.fai"
 
+// Collect GRIDSS reference files
+gridss_reference_files = Channel.fromPath( "${params.gridss_reference_fasta}.*", checkIfExists: true ).collect()
+
 workflow {
  /**
  * Validate the input bams
@@ -106,7 +118,6 @@ workflow {
  name: 'input_validation.txt',
  storeDir: "${params.output_dir_base}/validation/run_validate_PipeVal"
  )
-
  /**
  * Call "delly call -x hg19.excl -o t1.bcf -g hg19.fa tumor1.bam normal1.bam" per paired (tumor sample, normal sample)
  * The sv are stored in call_sSV_Delly.out.nt_call_bcf
@@ -158,7 +169,6 @@ workflow {
  call_sSV_Delly.out.nt_call_bcf_csi,
  call_sSV_Delly.out.tumor_id
  )
-
  /**
  * Filter the output bcf from filter_sSV_Delly.
  * The default filter_condition is "FILTER=='PASS'", which filters out NonPass calls.
@@ -168,7 +178,6 @@ workflow {
  params.filter_condition,
  call_sSV_Delly.out.tumor_id
  )
-
  /**
  * Generate one sha512 checksum for DELLY's output files.
  */
@@ -189,4 +198,12 @@ workflow {
  call_sSV_Manta.out.manta_vcfs.flatten()
  )
  }
+ if ('gridss2' in params.algorithm) {
+ preprocess_BAM_GRIDSS(
+ gridss_ch,
+ params.gridss_reference_fasta,
+ gridss_reference_files
+ )
+ }
+
  }

metadata.yaml

@@ -1,8 +1,8 @@
 ---
 category: "pipeline"
-description: "Nextflow pipeline to call somatic structural variants using DELLY and Manta"
+description: "Nextflow pipeline to call somatic structural variants using DELLY, Manta and GRIDSS2"
 maintainers: "Boutros Lab Infrastructure <BoutrosLabInfrastructure@mednet.ucla.edu>"
 languages: ["Nextflow", "Docker"]
 dependencies: ["Java", "Nextflow", "Docker"]
 references: "https://uclahs.box.com/s/qfzr99sc8ntmfddn30ii62wx4273utoz"
-tools: ["Delly:v1.2.6", "Manta:v1.6.0", "BCFtools:v1.15.1", "PipeVal:v4.0.0-rc.2"]
+tools: ["Delly:v1.2.6", "Manta:v1.6.0", "GRIDSS2:v2.13.2", "BCFtools:v1.15.1", "PipeVal:v4.0.0-rc.2"]

module/gridss.nf

@@ -0,0 +1,58 @@
+#!/usr/bin/env nextflow
+
+log.info """\
+------------------------------------
+ G R I D S S 2
+------------------------------------
+Docker Images:
+- docker_image_gridss: ${params.docker_image_gridss}
+"""
+
+include { generate_standard_filename; sanitize_string } from '../external/pipeline-Nextflow-module/modules/common/generate_standardized_filename/main.nf'
+
+process preprocess_BAM_GRIDSS {
+ container params.docker_image_gridss
+
+ publishDir "${params.workflow_output_dir}/intermediate/${task.process.replace(':', '/')}",
+ pattern: "${bam_name}.gridss.working/*",
+ mode: "copy",
+ saveAs: {
+ "${output_filename}.gridss.working/${output_filename}.${sanitize_string(file(it).getName().replace("${bam_name}.", ""))}"
+ }
+
+ publishDir "${params.log_output_dir}/process-log",
+ pattern: ".command.*",
+ mode: "copy",
+ saveAs: { "${task.process.replace(':', '/')}/${task.process}-${task.index}/log${file(it).getName()}" }
+
+ input:
+ tuple(val(sample_id), path(sample_bam), path(sample_index))
+ path(gridss_reference_fasta)
+ path(gridss_reference_files)
+
+ output:
+ path "${bam_name}.gridss.working/*", emit: gridss_preprocess
+ path ".command.*"
+
+ script:
+ gridss_mem = "${task.memory.toGiga()}g"
+ gridss_jar = "/usr/local/share/gridss-${params.gridss_version}-1/gridss.jar"
+ bam_name = file(sample_bam).getName()
+ output_filename = generate_standard_filename(
+ "GRIDSS2-${params.gridss_version}",
+ params.dataset_id,
+ sample_id,
+ [:]
+ )
+
+ """
+ set -euo pipefail
+ gridss \
+ -r ${gridss_reference_fasta} \
+ -j ${gridss_jar} \
+ -s preprocess \
+ -t ${task.cpus} \
+ --jvmheap ${gridss_mem} \
+ ${sample_bam}
+ """
+ }