From 5bbf9ae40b10a84bbfd633fa8ee3e10ce6869827 Mon Sep 17 00:00:00 2001
From: Laurent Modolo <laurent.modolo@ens-lyon.fr>
Date: Mon, 15 Jul 2024 15:24:39 +0200
Subject: [PATCH] add option to use paraplit instead of cutsite
---
conf/modules.config | 15 +++++++++++++++
modules/local/parasplit.nf | 28 ++++++++++++++++++++++++++++
nextflow.config | 4 ++++
nextflow_schema.json | 13 +++++++++++++
workflows/hic.nf | 19 +++++++++++++++++++
5 files changed, 79 insertions(+)
create mode 100644 modules/local/parasplit.nf
diff --git a/conf/modules.config b/conf/modules.config
index 08075e7..1d85bb1 100644
--- a/conf/modules.config
+++ b/conf/modules.config
@@ -470,6 +470,21 @@ process {
]
}
+ withName: 'PARASPLIT' {
+ ext.output_for = { "${meta1.id}_${meta1.chunk}_${meta1.mates}_digested.fastq" }
+ ext.output_rev = { "${meta2.id}_${meta2.chunk}_${meta2.mates}_digested.fastq" }
+ ext.args = { [
+ " --mode ${params.parasplit_mode}",
+ " --seed_size ${params.parasplit_seed}"
+ ].join('').trim()
+ }
+ publishDir = [
+ path: { "${params.outdir}/cutsite/digested"},
+ mode: 'copy',
+ enabled: params.save_digested
+ ]
+ }
+
withName: 'ITERALIGN' {
ext.prefix = { "${meta.id}_${meta.chunk}_${meta.mates}" }
ext.args = { [
diff --git a/modules/local/parasplit.nf b/modules/local/parasplit.nf
new file mode 100644
index 0000000..4bfb734
--- /dev/null
+++ b/modules/local/parasplit.nf
@@ -0,0 +1,28 @@
+process PARASPLIT {
+ tag "$meta1.id"
+ label 'process_high'
+
+ container = "docker.io/lbmc/parasplit:0.3.1"
+
+ input:
+ tuple val(meta1), path(reads1), val(meta2), path(reads2)
+ val(digestion)
+
+ output:
+ tuple val(meta1), path ("*.fastq"), emit: fastq
+ path "versions.yml", emit: versions
+
+ script:
+ def args = task.ext.args ?: ''
+ def output_for = task.ext.output_for ?: "${meta1.id}_${meta1.chunk}_${meta1.mates}.fastq"
+ def output_rev = task.ext.output_rev ?: "${meta2.id}_${meta2.chunk}_${meta2.mates}.fastq"
+
+ """
+ parasplit -sf ${reads1} -sr ${reads2} -le ${digestion} -of ${output_for} -or ${output_rev} -nt ${task.cpus} -sz 20 -m all ${args}
+ cat <<-END_VERSIONS > versions.yml
+ "${task.process}":
+ parasplit: v0.3.1)
+ END_VERSIONS
+ """
+}
+
diff --git a/nextflow.config b/nextflow.config
index 26dee50..eebf741 100644
--- a/nextflow.config
+++ b/nextflow.config
@@ -176,6 +176,10 @@ params {
save_digested = false
cutsite = false
+ //Parasplit
+ parasplit_mode = 'all'
+ parasplit_seed = 20
+
//Iterative alignement
iteralign = false
}
diff --git a/nextflow_schema.json b/nextflow_schema.json
index d26c093..bcc53ce 100644
--- a/nextflow_schema.json
+++ b/nextflow_schema.json
@@ -720,6 +720,19 @@
"description": "Mode to use to make the digestion.",
"enum": ["all", "for_vs_rev", "pile"]
},
+ "parasplit": {
+ "type": "boolean",
+ "description": "If true, do parasplit process to pre digest reads"
+ },
+ "parasplit_seed": {
+ "type": "integer",
+ "description": "Minimum size of a fragment (i.e. seed size used in mapping as reads smaller won't be mapped.)"
+ },
+ "parasplit_mode": {
+ "type": "string",
+ "description": "Mode to use to make the digestion.",
+ "enum": ["all", "fr"]
+ },
"iteralign": {
"type": "boolean",
"description": "If true, do iterative alignment"
diff --git a/workflows/hic.nf b/workflows/hic.nf
index 6c3487e..c2adb4b 100644
--- a/workflows/hic.nf
+++ b/workflows/hic.nf
@@ -127,6 +127,7 @@ include { TADS } from '../subworkflows/local/tads'
include { FASTQC } from '../modules/nf-core/fastqc/main'
include { CUSTOM_DUMPSOFTWAREVERSIONS } from '../modules/nf-core/custom/dumpsoftwareversions/main'
include { CUTSITE } from '../modules/local/hicstuff/cutsite'
+include { PARASPLIT } from '../modules/local/parasplit'
/*
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@@ -208,6 +209,24 @@ workflow HIC {
ch_reads = CUTSITE.out.fastq
ch_versions = ch_versions.mix(CUTSITE.out.versions)
}
+ if (params.parasplit){
+ // Align each mates separetly and add mates information in [meta]
+ ch_reads_r1 = ch_reads.map{ it -> pairToSingle(it,"R1") }
+ ch_reads_r2 = ch_reads.map{ it -> pairToSingle(it,"R2") }
+ ch_reads = ch_reads_r1.concat(ch_reads_r2)
+
+ ch_reads.combine(ch_reads)
+ .map {
+ meta1, reads1, meta2, reads2 ->
+ meta1.id == meta2.id && meta1.chunk == meta2.chunk && meta1.mates == "R1" && meta2.mates == "R2" ? [ meta1, reads1, meta2, reads2 ] : null
+ }.set{ new_ch_reads }
+ PARASPLIT(
+ new_ch_reads,
+ params.digestion
+ )
+ ch_reads = PARASPLIT.out.fastq
+ ch_versions = ch_versions.mix(PARASPLIT.out.versions)
+ }
//
// SUB-WORFLOW: HiC-Pro
--
GitLab