1
0
Fork 0
mirror of https://github.com/MillironX/taxprofiler.git synced 2024-11-21 19:06:03 +00:00

Addd FASTP adadapter list

This commit is contained in:
James Fellows Yates 2022-11-02 12:39:34 +01:00
parent 10f7f9e5de
commit d487a1ff2f
8 changed files with 26 additions and 13 deletions

View file

@ -62,7 +62,7 @@ process {
ext.args = [
// trimming options
params.shortread_qc_skipadaptertrim ? "--disable_adapter_trimming" : "",
params.shortread_qc_adapter1 ? "--adapter_sequence ${params.shortread_qc_adapter1}" : "",
params.shortread_qc_adapterlist ? "" : params.shortread_qc_adapter1 ? "--adapter_sequence ${params.shortread_qc_adapter1}" : "",
// filtering options
"--length_required ${params.shortread_qc_minlength}",
(params.perform_shortread_complexityfilter && params.shortread_complexityfilter_tool == 'fastp') ? "--low_complexity_filter --complexity_threshold ${params.shortread_complexityfilter_fastp_threshold}" : ''
@ -82,8 +82,8 @@ process {
params.shortread_qc_excludeunmerged ? '' : "--include_unmerged",
// trimming options
params.shortread_qc_skipadaptertrim ? "--disable_adapter_trimming" : "",
params.shortread_qc_adapter1 ? "--adapter_sequence ${params.shortread_qc_adapter1}" : "",
params.shortread_qc_adapter2 ? "--adapter_sequence_r2 ${params.shortread_qc_adapter2}" : "--detect_adapter_for_pe",
params.shortread_qc_adapterlist ? "" : params.shortread_qc_adapter1 ? "--adapter_sequence ${params.shortread_qc_adapter1}" : "",
params.shortread_qc_adapterlist ? "" : params.shortread_qc_adapter2 ? "--adapter_sequence_r2 ${params.shortread_qc_adapter2}" : "--detect_adapter_for_pe",
// filtering options
"--length_required ${params.shortread_qc_minlength}",
params.perform_shortread_complexityfilter && params.shortread_complexityfilter_tool == 'fastp' ? "--low_complexity_filter --complexity_threshold ${params.shortread_complexityfilter_fastp_threshold}" : ''

View file

@ -55,7 +55,7 @@
},
"fastp": {
"branch": "master",
"git_sha": "5e34754d42cd2d5d248ca8673c0a53cdf5624905"
"git_sha": "1e49f31e93c56a3832833eef90a02d3cde5a3f7e"
},
"fastqc": {
"branch": "master",

View file

@ -9,6 +9,7 @@ process FASTP {
input:
tuple val(meta), path(reads)
path adapter_fasta
val save_trimmed_fail
val save_merged
@ -27,6 +28,7 @@ process FASTP {
script:
def args = task.ext.args ?: ''
def prefix = task.ext.prefix ?: "${meta.id}"
def adapter_list = adapter_fasta ? "--adapter_fasta ${adapter_fasta}" : ""
def fail_fastq = save_trimmed_fail && meta.single_end ? "--failed_out ${prefix}.fail.fastq.gz" : save_trimmed_fail && !meta.single_end ? "--unpaired1 ${prefix}_1.fail.fastq.gz --unpaired2 ${prefix}_2.fail.fastq.gz" : ''
// Added soft-links to original fastqs for consistent naming in MultiQC
// Use single ended for interleaved. Add --interleaved_in in config.
@ -40,6 +42,7 @@ process FASTP {
--thread $task.cpus \\
--json ${prefix}.fastp.json \\
--html ${prefix}.fastp.html \\
$adapter_list \\
$fail_fastq \\
$args \\
2> ${prefix}.fastp.log \\
@ -61,6 +64,7 @@ process FASTP {
--thread $task.cpus \\
--json ${prefix}.fastp.json \\
--html ${prefix}.fastp.html \\
$adapter_list \\
$fail_fastq \\
$args \\
2> ${prefix}.fastp.log
@ -82,6 +86,7 @@ process FASTP {
--out2 ${prefix}_2.fastp.fastq.gz \\
--json ${prefix}.fastp.json \\
--html ${prefix}.fastp.html \\
$adapter_list \\
$fail_fastq \\
$merge_fastq \\
--thread $task.cpus \\

View file

@ -23,6 +23,10 @@ input:
List of input FastQ files of size 1 and 2 for single-end and paired-end data,
respectively. If you wish to run interleaved paired-end data, supply as single-end data
but with `--interleaved_in` in your `modules.conf`'s `ext.args` for the module.
- adapter_fasta:
type: file
description: File in FASTA format containing possible adapters to remove.
pattern: "*.{fasta,fna,fas,fa}"
- save_trimmed_fail:
type: boolean
description: Specify true to save files that failed to pass trimming thresholds ending in `*.fail.fastq.gz`

View file

@ -107,8 +107,8 @@
"shortread_qc_adapterlist": {
"type": "string",
"default": "None",
"fa_icon": "fas fa-list-ul",
"help_text": "This affects AdapterRemoval. It will replace default adapters or whatever is supplied in `--shortread_qc_adapter{1,2}`.\n\nThis allows you to mix libraries built with different adapters into one run.\n\n> Modifies tool parameter(s):\n> - AdapterRemoval: `--adapter-list`."
"description": "Specify a list of all possible adapters to trim. Overrides --shortread_qc_adapter1/2. Formats: .txt (AdapterRemoval) or .fasta. (fastp).",
"help_text": "Allows to supply a file with a list of adapter (combinations) to remove from all files. \n\nOverrides the --shortread_qc_adapter1/--shortread_qc_adapter2 parameters . \n\nFor AdapterRemoval this consists of a two column table with a `.txt` extension: first column represents forward strand, second column for reverse strand. You must supply all possibly combinations, one per line, and this list is applied to all files. See AdapterRemoval documentation for more information.\n\nFor fastp this consists of a standard FASTA format with a `.fasta`/`.fa`/`.fna`/`.fas` extension. The adapter sequence in this file should be at least 6bp long, otherwise it will be skipped. fastp trims the adapters present in the FASTA file one by one.\n\n> Modifies AdapterRemoval parameter: --adapter-list\n> Modifies fastp parameter: --adapter_fasta"
},
"shortread_qc_mergepairs": {
"type": "boolean",

View file

@ -8,6 +8,7 @@ include { FASTP as FASTP_PAIRED } from '../../modules/nf-core/fastp/main'
workflow SHORTREAD_FASTP {
take:
reads // [[meta], [reads]]
adapterlist
main:
ch_versions = Channel.empty()
@ -19,9 +20,9 @@ workflow SHORTREAD_FASTP {
paired: it[0]['single_end'] == false
}
FASTP_SINGLE ( ch_input_for_fastp.single, false, false )
FASTP_SINGLE ( ch_input_for_fastp.single, adapterlist, false, false )
// Last parameter here turns on merging of PE data
FASTP_PAIRED ( ch_input_for_fastp.paired, false, params.shortread_qc_mergepairs )
FASTP_PAIRED ( ch_input_for_fastp.paired, adapterlist, false, params.shortread_qc_mergepairs )
if ( params.shortread_qc_mergepairs ) {
ch_fastp_reads_prepped_pe = FASTP_PAIRED.out.reads_merged

View file

@ -18,7 +18,7 @@ workflow SHORTREAD_PREPROCESSING {
ch_multiqc_files = Channel.empty()
if ( params.shortread_qc_tool == "fastp" ) {
ch_processed_reads = SHORTREAD_FASTP ( reads ).reads
ch_processed_reads = SHORTREAD_FASTP ( reads, adapterlist ).reads
ch_versions = ch_versions.mix( SHORTREAD_FASTP.out.versions )
ch_multiqc_files = ch_multiqc_files.mix( SHORTREAD_FASTP.out.mqc )
} else if ( params.shortread_qc_tool == "adapterremoval" ) {

View file

@ -103,7 +103,12 @@ workflow TAXPROFILER {
ch_versions = Channel.empty()
ch_multiqc_logo= Channel.fromPath("$projectDir/docs/images/nf-core-taxprofiler_logo_custom_light.png")
ch_adapterlist_for_shortreadqc = params.shortread_qc_adapterlist ? file(params.shortread_qc_adapterlist) : []
adapterlist = params.shortread_qc_adapterlist ? file(params.shortread_qc_adapterlist) : []
if ( params.shortread_qc_adapterlist ) {
if ( params.shortread_qc_tool == 'adapterremoval' && !(adapterlist.extension == 'txt') ) error "[nf-core/taxprofiler] ERROR: AdapterRemoval2 adapter list requires a `.txt` format and extension. Check input: --shortread_qc_adapterlist ${params.shortread_qc_adapterlist}"
if ( params.shortread_qc_tool == 'fastp' && !adapterlist.extension.matches(".*(fa|fasta|fna|fas)") ) error "[nf-core/taxprofiler] ERROR: fastp adapter list requires a `.fasta` format and extension (or fa, fas, fna). Check input: --shortread_qc_adapterlist ${params.shortread_qc_adapterlist}"
}
/*
SUBWORKFLOW: Read in samplesheet, validate and stage input files
@ -134,10 +139,8 @@ workflow TAXPROFILER {
SUBWORKFLOW: PERFORM PREPROCESSING
*/
ch_adapterlist_for_shortreadqc
if ( params.perform_shortread_qc ) {
ch_shortreads_preprocessed = SHORTREAD_PREPROCESSING ( INPUT_CHECK.out.fastq, ch_adapterlist_for_shortreadqc ).reads
ch_shortreads_preprocessed = SHORTREAD_PREPROCESSING ( INPUT_CHECK.out.fastq, adapterlist ).reads
ch_versions = ch_versions.mix( SHORTREAD_PREPROCESSING.out.versions )
} else {
ch_shortreads_preprocessed = INPUT_CHECK.out.fastq