From e5216e2fb7d5139a11da997fb53aa70a6173da38 Mon Sep 17 00:00:00 2001
From: nservant <nicolas.servant@curie.fr>
Date: Wed, 2 Sep 2020 22:02:01 +0200
Subject: [PATCH] add --fastq_chunks_size parameter
---
CHANGELOG.md | 1 +
bin/merge_statfiles.py | 2 +-
docs/usage.md | 38 +++++++++++++++++++++-----------------
main.nf | 32 +++++++++++++++++---------------
nextflow.config | 11 ++++++-----
nextflow_schema.json | 42 ++++++++++++++++++++++++------------------
6 files changed, 70 insertions(+), 56 deletions(-)
diff --git a/CHANGELOG.md b/CHANGELOG.md
index b72dc20..719ea08 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -8,6 +8,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
### `Added`
* Template update for nf-core/tools v1.10.2
+* Add the `--fastq_chunks_size` to specify the number of reads per chunks if split_fastq is true
### `Fixed`
diff --git a/bin/merge_statfiles.py b/bin/merge_statfiles.py
index 469cacd..dc11bf7 100755
--- a/bin/merge_statfiles.py
+++ b/bin/merge_statfiles.py
@@ -48,7 +48,7 @@ if __name__ == "__main__":
if not line.startswith("#"):
lsp = line.strip().split("\t")
data = map(num, lsp[1:len(lsp)])
- template[str(lsp[0])] = data
+ template[str(lsp[0])] = list(data)
if len(template) == 0:
print("Cannot find template files !")
diff --git a/docs/usage.md b/docs/usage.md
index d4af9fd..11b0653 100644
--- a/docs/usage.md
+++ b/docs/usage.md
@@ -517,75 +517,79 @@ normalization. Default: 0.1
## Inputs/Outputs
-### `--splitFastq`
+### `--split_fastq`
By default, the nf-core Hi-C pipeline expects one read pairs per sample.
However, for large Hi-C data processing single fastq files can be very
time consuming.
-The `--splitFastq` option allows to automatically split input read pairs
-into chunks of reads. In this case, all chunks will be processed in parallel
+The `--split_fastq` option allows to automatically split input read pairs
+into chunks of reads of size `--fastq_chunks_size` (Default : 20000000). In this case, all chunks will be processed in parallel
and merged before generating the contact maps, thus leading to a significant
increase of processing performance.
```bash
---splitFastq '[Number of reads per chunk]'
+--split_fastq --fastq_chunks_size '[numeric]'
```
-### `--saveReference`
+### `--save_reference`
If specified, annotation files automatically generated from the `--fasta` file
are exported in the results folder. Default: false
```bash
---saveReference
+--save_reference
```
-### `--saveAlignedIntermediates`
+### `--save_aligned_intermediates`
If specified, all intermediate mapping files are saved and exported in the
results folder. Default: false
```bash
---saveReference
+--save_aligned_inermediates
```
-### `--saveInteractionBAM`
+### `--save_interaction_bam`
If specified, write a BAM file with all classified reads (valid paires,
dangling end, self-circle, etc.) and its tags.
+```bash
+--save_interaction_bam
+```
+
## Skip options
-### `--skipMaps`
+### `--skip_maps`
If defined, the workflow stops with the list of valid interactions, and the
genome-wide maps are not built. Usefult for capture-C analysis. Default: false
```bash
---skipMaps
+--skip_maps
```
-### `--skipIce`
+### `--skip_ice`
If defined, the ICE normalization is not run on the raw contact maps.
Default: false
```bash
---skipIce
+--skip_ice
```
-### `--skipCool`
+### `--skip_cool`
If defined, cooler files are not generated. Default: false
```bash
---skipCool
+--skip_cool
```
-### `--skipMultiQC`
+### `--skip_multiQC`
If defined, the MultiQC report is not generated. Default: false
```bash
---skipMultiQC
+--skip_multiQC
```
diff --git a/main.nf b/main.nf
index 5763152..495a1fe 100644
--- a/main.nf
+++ b/main.nf
@@ -34,7 +34,8 @@ def helpMessage() {
--save_reference [bool] Save reference genome to output folder. Default: False
Alignments
- --split_fastq [int] Size of read chuncks to use to speed up the workflow. Default: None
+ --split_fastq [bool] Split fastq files in reads chunks to speed up computation. Default: None
+ --fastq_chunks_size [int] Size of read chunks if split_fastq is true. Default: 20000000
--save_aligned_intermediates [bool] Save intermediates alignment files. Default: False
--bwt2_opts_end2end [str] Options for bowtie2 end-to-end mappinf (first mapping step). See hic.config for default.
--bwt2_opts_trimmed [str] Options for bowtie2 mapping after ligation site trimming. See hic.config for default.
@@ -46,14 +47,14 @@ def helpMessage() {
--rm_dup [bool] Remove duplicates. Default: true
Contacts calling
- --min_restriction_fragment_size [int] Minimum size of restriction fragments to consider. Default: None
- --max_restriction_fragment_size [int] Maximum size of restriction fragments to consider. Default: None
- --min_insert_size [int] Minimum insert size of mapped reads to consider. Default: None
- --max_insert_size [int] Maximum insert size of mapped reads to consider. Default: None
+ --min_restriction_fragment_size [int] Minimum size of restriction fragments to consider. Default: 0
+ --max_restriction_fragment_size [int] Maximum size of restriction fragments to consider. Default: 0
+ --min_insert_size [int] Minimum insert size of mapped reads to consider. Default: 0
+ --max_insert_size [int] Maximum insert size of mapped reads to consider. Default: 0
--save_interaction_bam [bool] Save BAM file with interaction tags (dangling-end, self-circle, etc.). Default: False
--dnase [bool] Run DNase Hi-C mode. All options related to restriction fragments are not considered. Default: False
- --min_cis_dist [int] Minimum intra-chromosomal distance to consider. Default: None
+ --min_cis_dist [int] Minimum intra-chromosomal distance to consider. Default: 0
Contact maps
--bin_size [int] Bin size for contact maps (comma separated). Default: '1000000,500000'
@@ -163,7 +164,7 @@ if (params.input_paths){
if ( params.split_fastq ){
raw_reads_full = raw_reads.concat( raw_reads_2 )
- raw_reads = raw_reads_full.splitFastq( by: params.split_fastq , file: true)
+ raw_reads = raw_reads_full.splitFastq( by: params.fastq_chunks_size, file: true)
}else{
raw_reads = raw_reads.concat( raw_reads_2 ).dump(tag: "data")
}
@@ -239,6 +240,8 @@ if(workflow.revision) summary['Pipeline Release'] = workflow.revision
summary['Run Name'] = custom_runName ?: workflow.runName
summary['Input'] = params.input
summary['splitFastq'] = params.split_fastq
+if (params.split_fastq)
+ summary['Read chunks Size'] = params.fastq_chunks_size
summary['Fasta Ref'] = params.fasta
summary['Restriction Motif']= params.restriction_site
summary['Ligation Motif'] = params.ligation_site
@@ -647,12 +650,12 @@ if (!params.dnase){
}
def opts = ""
- if ("$params.min_cis_dist".isInteger()) opts="${opts} -d ${params.min_cis_dist}"
- if ("$params.min_insert_size".isInteger()) opts="${opts} -s ${params.min_insert_size}"
- if ("$params.max_insert_size".isInteger()) opts="${opts} -l ${params.max_insert_size}"
- if ("$params.min_restriction_fragment_size".isInteger()) opts="${opts} -t ${params.min_restriction_fragment_size}"
- if ("$params.max_restriction_fragment_size".isInteger()) opts="${opts} -m ${params.max_restriction_fragment_size}"
- if (params.save_interaction_bam) opts="${opts} --sam"
+ opts += params.min_cis_dist > 0 ? " -d ${params.min_cis_dist}" : ''
+ opts += params.min_insert_size > 0 ? " -s ${params.min_insert_size}" : ''
+ opts += params.max_insert_size > 0 ? " -l ${params.max_insert_size}" : ''
+ opts += params.min_restriction_fragment_size > 0 ? " -t ${params.min_restriction_fragment_size}" : ''
+ opts += params.max_restriction_fragment_size > 0 ? " -m ${params.max_restriction_fragment_size}" : ''
+ opts += params.save_interaction_bam ? " --sam" : ''
prefix = pe_bam.toString() - ~/.bam/
"""
mapped_2hic_fragments.py -f ${frag_file} -r ${pe_bam} --all ${opts}
@@ -680,8 +683,7 @@ else{
sample = sample.toString() - ~/(\.[0-9]+)$/
}
- def opts = ""
- if ("$params.min_cis_dist".isInteger()) opts="${opts} -d ${params.min_cis_dist}"
+ opts = params.min_cis_dist > 0 ? " -d ${params.min_cis_dist}" : ''
prefix = pe_bam.toString() - ~/.bam/
"""
mapped_2hic_dnase.py -r ${pe_bam} ${opts}
diff --git a/nextflow.config b/nextflow.config
index edb8038..c765a4a 100644
--- a/nextflow.config
+++ b/nextflow.config
@@ -17,6 +17,7 @@ params {
genome = false
input_paths = false
split_fastq = false
+ fastq_chunks_size = 20000000
chromosome_size = false
restriction_fragments = false
skip_maps = false
@@ -34,12 +35,12 @@ params {
// Digestion Hi-C
restriction_site = 'A^AGCTT'
ligation_site = 'AAGCTAGCTT'
- min_restriction_fragment_size = false
- max_restriction_fragment_size = false
- min_insert_size = false
- max_insert_size = false
+ min_restriction_fragment_size = 0
+ max_restriction_fragment_size = 0
+ min_insert_size = 0
+ max_insert_size = 0
dnase = false
- min_cis_dist = false
+ min_cis_dist = 0
rm_dup = true
rm_singleton = true
rm_multi = true
diff --git a/nextflow_schema.json b/nextflow_schema.json
index ed2f701..9071bd2 100644
--- a/nextflow_schema.json
+++ b/nextflow_schema.json
@@ -27,10 +27,16 @@
"default": "undefined"
},
"split_fastq": {
- "type": "number",
- "description": "Split the reads into chunks before running. Specify the number of reads per chuncks as --split_fastq 20000000.",
- "fa_icon": "fas fa-dna"
+ "type": "boolean",
+ "description": "Split the reads into chunks before running the pipelne",
+ "fa_icon": "fas fa-dna",
+ "default": "false"
},
+ "fastq_chunks_size":{
+ "type": "integer",
+ "description": "Read number per chunks if split_fastq is used",
+ "default": "20000000"
+ },
"single_end": {
"type": "boolean",
"description": "Specifies that the input is single-end reads.",
@@ -178,29 +184,29 @@
"fa_icon": "fas fa-signature",
"properties": {
"min_cis_dist": {
- "type": "string",
- "default": "undefined",
- "description": "Minimum distance between loci to consider. Useful for --dnase mode to remove spurious ligation products"
+ "type": "integer",
+ "default": "O",
+ "description": "Minimum distance between loci to consider. Useful for --dnase mode to remove spurious ligation products. Only values > 0 are considered"
},
"max_insert_size": {
- "type": "string",
- "default": "undefined",
- "description": "Maximum fragment size to consider"
+ "type": "integer",
+ "default": "0",
+ "description": "Maximum fragment size to consider. Only values > 0 are considered"
},
"min_insert_size": {
- "type": "string",
- "default": "undefined",
- "description": "Minimum fragment size to consider"
+ "type": "integer",
+ "default": "0",
+ "description": "Minimum fragment size to consider. Only values > 0 are considered"
},
"max_restriction_fragment_size": {
- "type": "string",
- "default": "undefined",
- "description": "Maximum restriction fragment size to consider"
+ "type": "integer",
+ "default": "0",
+ "description": "Maximum restriction fragment size to consider. Only values > 0 are considered"
},
"min_restriction_fragment_size": {
- "type": "string",
- "default": "undefined",
- "description": "Minimum restriction fragment size to consider"
+ "type": "integer",
+ "default": "0",
+ "description": "Minimum restriction fragment size to consider. Only values > 0 are considered"
}
}
},
--
GitLab