Merge pull request #251 from replikation/feature_param-artic-normalize

[Feature] added parameter artic_normalize
replikation · Aug 3, 2023 · aa993eb · aa993eb
2 parents 19b87bf + 0854b0e
commit aa993eb
Show file tree

Hide file tree

Showing 4 changed files with 31 additions and 12 deletions.
diff --git a/nextflow.config b/nextflow.config
@@ -20,6 +20,8 @@ params {
     list = false
     samples = false
 
+    artic_normalize = 500
+
     // consensus qc
     reference_for_qc = ''
     seq_threshold = '0.90'

diff --git a/poreCov.nf b/poreCov.nf
@@ -347,11 +347,11 @@ workflow {
 
             // use medaka or nanopolish artic reconstruction
             if (params.nanopolish) { 
-                artic_ncov_np_wf(filtered_reads_ch, dir_input_ch, basecalling_wf.out[1])
+                artic_ncov_np_wf(filtered_reads_ch, dir_input_ch, basecalling_wf.out[1], artic_ncov_np_wf)
                 fasta_input_ch = artic_ncov_np_wf.out[0]
                 }
             else if (!params.nanopolish) { 
-                artic_ncov_wf(filtered_reads_ch) 
+                artic_ncov_wf(filtered_reads_ch, params.artic_normalize) 
                 fasta_input_ch = artic_ncov_wf.out[0] 
                 }
         }
@@ -379,11 +379,11 @@ workflow {
 
                 external_primer_schemes = Channel.fromPath(workflow.projectDir + "/data/external_primer_schemes", checkIfExists: true, type: 'dir' )
 
-                artic_ncov_np_wf(filtered_reads_ch, dir_input_ch, sequence_summary_ch )
+                artic_ncov_np_wf(filtered_reads_ch, dir_input_ch, sequence_summary_ch, artic_ncov_np_wf)
                 fasta_input_ch = artic_ncov_np_wf.out
                 }
             else if (!params.nanopolish) { 
-                artic_ncov_wf(filtered_reads_ch)
+                artic_ncov_wf(filtered_reads_ch, params.artic_normalize)
                 fasta_input_ch = artic_ncov_wf.out
                 }
         }
@@ -466,6 +466,9 @@ ${c_yellow}Inputs (choose one):${c_reset}
                     ${c_dim}[Lineage + Reports]${c_reset}
 
 ${c_yellow}Workflow control (optional)${c_reset}
+    --artic_normalize        Normalise down to moderate coverage to save runtime [default: $params.artic_normalize]
+                                   ${c_dim}(after mapping and before variant calling in the ARTIC bioinformatics pipeline)
+                                   Use `--artic_normalize False` to turn off this normalisation.${c_reset}
     --update                 Always try to use latest pangolin & nextclade release [default: $params.update]
     --samples                .csv input (header: Status,_id), renames barcodes (Status) by name (_id), e.g.:
                              Status,_id

diff --git a/workflows/artic_nanopore_nCov19.nf b/workflows/artic_nanopore_nCov19.nf
@@ -4,14 +4,15 @@ include { covarplot; covarplot_custom_bed } from './process/covarplot.nf'
 workflow artic_ncov_wf {
     take:   
         fastq
+        normalise_threshold
     main: 
 
         // assembly with a primer bed file
         if (params.primerV.toString().contains(".bed")) {
             primerBed = Channel.fromPath(params.primerV, checkIfExists: true )
             external_primer_schemes = Channel.fromPath(workflow.projectDir + "/data/external_primer_schemes", checkIfExists: true, type: 'dir' )
 
-            artic_medaka_custom_bed(fastq.combine(external_primer_schemes).combine(primerBed))
+            artic_medaka_custom_bed(fastq.combine(external_primer_schemes).combine(primerBed), normalise_threshold)
             assembly = artic_medaka_custom_bed.out.fasta
 
             // plot amplicon coverage
@@ -22,7 +23,7 @@ workflow artic_ncov_wf {
         else {
             external_primer_schemes = Channel.fromPath(workflow.projectDir + "/data/external_primer_schemes", checkIfExists: true, type: 'dir' )
 
-            artic_medaka(fastq.combine(external_primer_schemes))
+            artic_medaka(fastq.combine(external_primer_schemes), normalise_threshold)
             assembly = artic_medaka.out.fasta
 
             // plot amplicon coverage
@@ -42,6 +43,7 @@ workflow artic_ncov_np_wf {
         fastq
         fast5
         sequence_summaries
+        normalise_threshold
     main: 
 
         // assembly
@@ -55,7 +57,8 @@ workflow artic_ncov_np_wf {
                     .combine(fast5.map{it -> it[1]})
                     .combine(sequence_summaries)
                     .combine(primerBed)
-                    .map{it -> tuple(it[0],it[1],it[2],it[3],it[5],it[6])}
+                    .map{it -> tuple(it[0],it[1],it[2],it[3],it[5],it[6])},
+                normalise_threshold
         )
 
             assembly = artic_nanopolish_custom_bed.out.fasta
@@ -73,7 +76,8 @@ workflow artic_ncov_np_wf {
                         .combine(external_primer_schemes)
                         .combine(fast5.map{it -> it[1]})
                         .combine(sequence_summaries)
-                        .map{it -> tuple(it[0],it[1],it[2],it[3],it[5])}
+                        .map{it -> tuple(it[0],it[1],it[2],it[3],it[5])},
+                    normalise_threshold
             )
 
             assembly = artic_nanopolish.out.fasta

diff --git a/workflows/process/artic.nf b/workflows/process/artic.nf
@@ -11,6 +11,7 @@ process artic_medaka {
 
     input:
         tuple val(name), path(reads), path(external_scheme)
+        val(normalise_threshold)
     output:
         tuple val(name), path("*.consensus.fasta"), emit: fasta
         tuple val(name), path("${name}_mapped_*.primertrimmed.sorted.bam"), path("${name}_mapped_*.primertrimmed.sorted.bam.bai"), emit: reference_bam
@@ -22,11 +23,12 @@ process artic_medaka {
         tuple val(name), path("${name}.fail.vcf"), emit: vcf_fail
 
     script:   
+        def normalise_arg = normalise_threshold ? "--normalise ${normalise_threshold}" : ''
         """
         artic minion    --medaka \
                         --medaka-model ${params.medaka_model} \
                         --min-depth ${params.min_depth} \
-                        --normalise 500 \
+                        ${normalise_arg} \
                         --threads ${task.cpus} \
                         --scheme-directory ${external_scheme} \
                         --read-file ${reads} \
@@ -76,6 +78,7 @@ process artic_medaka_custom_bed {
 
     input:
         tuple val(name), path(reads), path(external_scheme), path(primerBed)
+        val(normalize_threshold)
     output:
         tuple val(name), path("*.consensus.fasta"), emit: fasta
         tuple val(name), path("${name}_mapped_*.primertrimmed.sorted.bam"), path("${name}_mapped_*.primertrimmed.sorted.bam.bai"), emit: reference_bam
@@ -86,6 +89,7 @@ process artic_medaka_custom_bed {
         tuple val(name), path("${name}.coverage_mask.txt"), emit: coverage_mask
         tuple val(name), path("${name}.fail.vcf"), emit: vcf_fail
     script:   
+        def normalise_arg = normalise_threshold ? "--normalise ${normalise_threshold}" : ''
         """
         # create a new primer dir as input for artic
         mkdir -p primer_scheme/nCoV-2019
@@ -101,7 +105,7 @@ process artic_medaka_custom_bed {
         artic minion    --medaka \
                         --medaka-model ${params.medaka_model} \
                         --min-depth ${params.min_depth} \
-                        --normalise 500 \
+                        ${normalise_arg} \
                         --threads ${task.cpus} \
                         --scheme-directory primer_scheme \
                         --read-file ${reads} \
@@ -152,6 +156,7 @@ process artic_nanopolish {
 
     input:
         tuple val(name), path(reads), path(external_scheme), path(fast5_dir), path(txt_files)
+        val(normalise_threshold)
     output:
         tuple val(name), path("*.consensus.fasta"), emit: fasta
         tuple val(name), path("${name}_mapped_*.primertrimmed.sorted.bam"), path("${name}_mapped_*.primertrimmed.sorted.bam.bai"), emit: reference_bam
@@ -162,8 +167,10 @@ process artic_nanopolish {
         tuple val(name), path("${name}.coverage_mask.txt"), emit: coverage_mask
         tuple val(name), path("${name}.fail.vcf"), emit: vcf_fail
     script:   
+        def normalise_arg = normalise_threshold ? "--normalise ${normalise_threshold}" : ''
         """
-        artic minion --minimap2 --normalise 500 \
+        artic minion --minimap2 \
+            ${normalise_arg} \
             --threads ${task.cpus} \
             --scheme-directory ${external_scheme} \
             --read-file ${reads} \
@@ -216,6 +223,7 @@ process artic_nanopolish_custom_bed {
 
     input:
         tuple val(name), path(reads), path(external_scheme), path(fast5_dir), path(txt_files), path(primerBed)
+        val(normalise_threshold)
     output:
         tuple val(name), path("*.consensus.fasta"), emit: fasta
         tuple val(name), path("${name}_mapped_*.primertrimmed.sorted.bam"), path("${name}_mapped_*.primertrimmed.sorted.bam.bai"), emit: reference_bam
@@ -226,6 +234,7 @@ process artic_nanopolish_custom_bed {
         tuple val(name), path("${name}.coverage_mask.txt"), emit: coverage_mask
         tuple val(name), path("${name}.fail.vcf"), emit: vcf_fail
     script:   
+        def normalise_arg = normalise_threshold ? "--normalise ${normalise_threshold}" : ''
         """
         # create a new primer dir as input for artic
         mkdir -p primer_scheme/nCoV-2019
@@ -238,7 +247,8 @@ process artic_nanopolish_custom_bed {
             sort -k4 > primer_scheme/nCoV-2019/V_custom/nCoV-2019.scheme.bed
 
         # start artic
-        artic minion --minimap2 --normalise 500 \
+        artic minion --minimap2 \
+            ${normalise_arg} \
             --threads ${task.cpus} \
             --scheme-directory primer_scheme \
             --read-file ${reads} \