first working version

config/compute_resources.config

@@ -7,10 +7,10 @@ executor {
 
 process {
     cpus = 1
-    maxForks = 20
+    maxForks = 4
     shell = ['/bin/bash', '-euo', 'pipefail']
     stageOutMode = 'rsync'
-    withName: 'agat' {
+    withLabel: 'agat' {
         cpus = 1
     }
     withName: 'prokka' {

config/software_packages.config

 process {
 
-    withName: 'agat' {
+    withLabel: 'agat' {
         conda = { singularity.enabled || docker.enabled ? '' : "$baseDir/conda/process_fastp.yml" }
         container = 'quay.io/biocontainers/agat:1.0.0--pl5321hdfd78af_0'
     }

main.nf

 #! /usr/bin/env nextflow
-
+/*
+ *  Jacques Dainat 2023
+ */
 nextflow.enable.dsl=2
 
 /*
@@ -15,6 +17,7 @@ params.species = "coli"
 params.strain = "POO247"
 params.codon_table = 11
 params.evalue = 1e-9
+params.output = "results"
 
 params.bakta_db="/path/to/bakta_db" // optional
 params.busco_db = "/path/to/busco_db" // optional
@@ -34,10 +37,10 @@ if(params.genus){
 }
 
 // Deal with BAKTA Parameters
-params.bakta = ' '
+params.bakta = ''
 File bakta_db_file = new File( params.bakta_db );
-if (bakta_db_file.exists() ){
-  params.bakta = '--db ${bakta_db}'
+if (!bakta_db_file.exists() ){
+  exit 1, "Bakta database name ${params.bakta_db} does not exists! It is mandatory if you wish to run Bakta. Otherwise deactivate bakta (--skip_bakta).\n"
 }
 
 // Deal with BUSCO Parameters
@@ -52,6 +55,10 @@ if(params.busco_lineage){
   params.busco = '${params.busco} --auto-lineage-prok'
 }
 
+// ****************************** MODULES ****************************
+include { busco_assembly; busco_annotation ; busco_annotation as busco_meta_annotation} from "${baseDir}/modules/busco.nf"
+include { agat_sp_extract_sequences; agat_sp_extract_sequences as extractMetaAA; agat_sp_merge_annotations } from "${baseDir}/modules/agat.nf"
+
 log.info """
 IRD
 .-./`) .-------.     ______
@@ -76,21 +83,28 @@ MbaP - Mutualize Bacterial Predictors
 
  """
 
-/*if( !params.skip_trimming && !(params.trimmer in trimming_tools) ){
-    exit 1, "Error: ${params.trimmer} is not a valid trimming tool option.\n Please provide a valid option from ${trimming_tools}.\n"
-}
-if( !params.skip_trimming && params.trimmer == 'trimmomatic' && !file(params.trimmomatic_adapter_path).exists() ){
-    exit 1, "The adapter file '${params.trimmomatic_adapter_path}' does not exist!\n"
-}
-*/
-
 workflow {
 
     main:
-        Channel.fromPath(params.assembly, checkIfExists: true)
-            .ifEmpty { exit 1, "Cannot find assembly file matching ${params.assembly}!\n" }
-            .set {assembly}
-        mbap(assembly)
+        File input = new File( "${params.assembly}");
+        if ( input.exists() ){
+          // Folder case
+          if ( input.isDirectory()) {
+            Channel.fromPath( "${params.assembly}/*", type: 'file', checkIfExists: true)
+            .ifEmpty { exit 1, "No file within the ${params.assembly } folder found !\n" }
+            .set {assemblies}
+          }
+          // file case
+          else {
+            Channel.fromPath( params.assembly, type: 'file', checkIfExists: true)
+            .ifEmpty { exit 1, "No file with ${params.assembly } found !\n" }
+            .set {assemblies}
+          }
+        }
+        else {
+          exit 1, "${params.assembly} input does not exists!"
+        }
+        mbap(assemblies)
 }
 
 workflow mbap {
@@ -99,30 +113,71 @@ workflow mbap {
         assembly
 
     main:
-        //busco(assembly)
-        prokka(assembly)
-        bakta(assembly)
+        make_id(assembly)
+        make_id.out[0].set{id_assembly}
+        make_id.out[1].set{id_assembly_bis}
+        //id_assembly.view()
+        busco_assembly(id_assembly)
+        prokka(id_assembly)
+        prokka.out[0].set{annotations}
+        bakta(id_assembly)
+        annotations.concat(bakta.out[0]).set{annotations}
+        annotations.multiMap { it ->
+                               first: it
+                               second: it
+                             }.set{annotationsForked}
+        //annotations.view()
+        // annotations.concat(bakta.out[0]).groupTuple().set{annotations} // GOOD
+        id_assembly.combine(annotationsForked.first, by:0).set{id_assembly_annot}
+        //id_assembly.join(annotations).set{id_assembly_annot} // GOOD
+        //id_assembly_annot.view()
+        agat_sp_extract_sequences(id_assembly_annot).set{proteomes}
+        busco_annotation(proteomes)
+        //proteomes.map{fa, annotation -> [ fa.baseName, fa, annotation ]}.groupTuple().set{proteomesByAssembly}
+
+        annotationsForked.second.groupTuple().set{annotations_group}
+        id_assembly_bis.join(annotations_group).set{id_assembly_annots}
+        agat_sp_merge_annotations(id_assembly_annots)
+        extractMetaAA(agat_sp_merge_annotations.out)
+        busco_meta_annotation(extractMetaAA.out)
+}
+
+process make_id {
+
+    tag "${assembly}"
+
+    input:
+        path(assembly)
 
+    output:
+        tuple val("${assembly.baseName}"), path("${assembly}")
+        tuple val("${assembly.baseName}"), path("${assembly}")
+
+    script:
+    """
+    """
 }
 
+
 process prokka {
 
     tag "${assembly}"
-    publishDir "${params.outdir}/${assembly}/prokka", mode: 'copy'
+    publishDir "${params.outdir}/${id}/prokka", mode: 'copy'
 
     input:
-    path(assembly)
+        tuple val(id), path(assembly)
 
     output:
+    tuple val(id), file('*/*.gff*')
     file('*')
 
     script:
     """
     prokka \
-        --addgenes --locustag ${params.locus_prefix} \
-        --increment 10 --compliant \
+        --addgenes  --addmrna --locustag ${params.locus_prefix} \
+        --outdir ${id}_prokka --prefix ${id}_annotation_prokka \
         --kingdom Bacteria --gcode ${params.codon_table} \
-        --evalue ${params.evalue} --rfam \
+        --evalue ${params.evalue} --rfam --cpus ${task.cpus}\
         ${params.prokka} \
         ${assembly}
     """
@@ -132,40 +187,33 @@ process prokka {
 process bakta {
 
     tag "$assembly"
-    publishDir "${params.outdir}/${assembly}/bakta", mode: 'copy'
+    publishDir "${params.outdir}/${id}/bakta", mode: 'copy'
     conda 'bakta'
 
     input:
-    path(assembly)
+      tuple val(id), path(assembly)
 
     output:
-    path("${assembly}_bakta")
+      tuple val(id), file("${id}_annotation_bakta.gff")
+      path("${id}_bakta")
+
 
     script:
     """
-    bakta \
+      bakta \
       --keep-contig-headers --genus ${params.genus} \
-      --species ${params.species} --prefix ${assembly} --locus-tag ${params.locus_prefix} --output ${assembly}_bakta \
-      --verbose \
+      --species ${params.species} --prefix ${id} \
+      --locus-tag ${params.locus_prefix} --db ${params.bakta_db} \
+      --verbose --threads ${task.cpus} \
+      --output ${id}_bakta \
       ${params.bakta} \
       ${assembly}
+
+       mv ${id}_bakta/${id}.gff3 ${id}_annotation_bakta.gff
     """
 }
 
-process busco {
-    label 'busco'
-    publishDir "${params.output}/${assembly}/assembly", mode: 'copy'
-    input:
-        tuple val(id), path(contigs)
-        val(lineage)
-    output:
-        path("busco_${assembly}")
 
-    script:
-        """
-        busco -i ${assembly} -o busco_${assembly} --mode genome -f ${params.busco}
-        """
-}
 
 
 /**************         onComplete         ****************/

modules/agat.nf

+/* ****************** AGAT ****************** */
+
+process agat_sp_extract_sequences {
+    tag "${id}"
+    label 'agat'
+    publishDir "${params.output}/${assembly.baseName}/agat", mode: 'copy'
+    input:
+        tuple val(id), file(assembly), file(annotation)
+    output:
+        tuple val(id), file(assembly), file("${annotation.baseName}.fa")
+
+    script:
+
+        """
+        agat_sp_extract_sequences.pl --gff ${annotation} --fasta ${assembly} -p -o ${annotation.baseName}.fa
+        """
+}
+
+process agat_sp_merge_annotations {
+    tag "${id}"
+    label 'agat'
+    publishDir "${params.output}/${assembly.baseName}/agat", mode: 'copy'
+    input:
+        tuple val(id), file(assembly), file(annotations)
+    output:
+        tuple val(id), file(assembly), file("*_complemented.gff")
+
+    script:
+
+        def annotation_list = []
+        annotation_list = annotations
+        annotation_list_bash = annotation_list.join(" "); // remove bracket and replace comma by space to be processed by bash
+        """
+        parameters=""
+        for i in ${annotation_list_bash};do
+            parameters="\${parameters} --gff \${i}"
+        done
+        agat_sp_merge_annotations.pl \${parameters} -o ${id}_complemented.gff
+        """
+}

modules/busco.nf

+/* ****************** BUSCO ****************** */
+
+process busco_assembly {
+    tag "${id}"
+    label 'busco'
+    publishDir "${params.output}/${id}/busco_assembly", mode: 'copy'
+    input:
+        tuple val(id), path(assembly)
+    output:
+        path("busco_${id}")
+
+    script:
+        """
+        busco -i ${assembly} -o busco_${id} --cpu ${task.cpus} --mode genome -f ${params.busco}
+        """
+}
+
+process busco_annotation {
+    tag "${id}"
+    label 'busco'
+    publishDir "${params.output}/${id}/busco_annotation", mode: 'copy'
+    input:
+        tuple val(id), file(assembly), file(annotation)
+
+    output:
+        path("busco_${annotation_tool}")
+
+    script:
+        annotation_tool = annotation.toString().tokenize(".")[-2].tokenize("_")[-1]
+
+        """
+        busco -i ${annotation} -o busco_${annotation_tool} --cpu ${task.cpus} --mode protein -f ${params.busco}
+        """
+}

nextflow.config

@@ -27,6 +27,10 @@ profiles {
         includeConfig "$baseDir/config/software_packages.config"
     }
 
+    res_local {
+      includeConfig "$baseDir/config/compute_resources.config"
+    }
+
     itrop {
         executor {
             name = 'slurm'