Improve GUNC execution

nf-core · Nov 1, 2024 · a4f42ef · a4f42ef
1 parent da52285
commit a4f42ef
Show file tree

Hide file tree

Showing 2 changed files with 32 additions and 20 deletions.
diff --git a/modules/local/combine_tsv.nf b/modules/local/combine_tsv.nf
@@ -7,16 +7,16 @@ process COMBINE_TSV {
         'biocontainers/bioawk:1.0--hed695b0_5' }"
 
     input:
-    path(bin_summaries, stageAs: "bin_summaries/*.tsv")
+    path(tsv_files, stageAs: "tsv_files/*.tsv")
 
     output:
     path("*.tsv")      , emit: combined
     path "versions.yml", emit: versions
 
     script:
-    def prefix = task.ext.prefix ?: "bin_depths_summary_combined"
+    def prefix = task.ext.prefix ?: "tsv_summary_combined"
     """
-    bioawk '(NR == 1) || (FNR > 1)' ${bin_summaries} > ${prefix}.tsv
+    bioawk '(NR == 1) || (FNR > 1)' ${tsv_files} > ${prefix}.tsv
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":

diff --git a/subworkflows/local/bin_qc.nf b/subworkflows/local/bin_qc.nf
@@ -10,18 +10,19 @@ include { CHECKM_QA                         } from '../../modules/nf-core/checkm
 include { CHECKM_LINEAGEWF                  } from '../../modules/nf-core/checkm/lineagewf/main'
 include { CHECKM2_PREDICT                   } from '../../modules/nf-core/checkm2/predict/main'
 include { COMBINE_TSV as COMBINE_BINQC_TSV  } from '../../modules/local/combine_tsv'
+include { COMBINE_TSV as COMBINE_GUNC_TSV   } from '../../modules/local/combine_tsv'
 include { GUNC_DOWNLOADDB                   } from '../../modules/nf-core/gunc/downloaddb/main'
 include { GUNC_RUN                          } from '../../modules/nf-core/gunc/run/main'
 include { GUNC_MERGECHECKM                  } from '../../modules/nf-core/gunc/mergecheckm/main'
 
 
 workflow BIN_QC {
     take:
-    ch_bins        // [ [ meta] , fasta           ], input bins (mandatory)
-    ch_checkm_db   // [ db                        ], presupplied CheckM database (optional)
-    ch_checkm2_db  // [ [meta]  , db              ], presupplied CheckM2 database (optional)
-    ch_busco_db    // [ db                        ], presupplied BUSCO database (optional)
-    ch_gunc_db     // [ db                        ], presupplied GUNC database (optional)
+    ch_bins         // [ [ meta] , fasta           ], input bins (mandatory)
+    ch_checkm_db    // [ db                        ], presupplied CheckM database (optional)
+    ch_checkm2_db   // [ [meta]  , db              ], presupplied CheckM2 database (optional)
+    ch_busco_db     // [ db                        ], presupplied BUSCO database (optional)
+    ch_gunc_db      // [ db                        ], presupplied GUNC database (optional)
 
     main:
     qc_summary = []
@@ -81,7 +82,7 @@ workflow BIN_QC {
         )
 
         ch_multiqc_files = ch_multiqc_files.mix(
-            BUSCO.out.summary_domain.mix(BUSCO.out.summary_specific).map{ it[1] }
+            BUSCO.out.summary_domain.mix(BUSCO.out.summary_specific).map { it[1] }
         )
         qc_summary = BUSCO_SUMMARY.out.summary
         ch_versions = ch_versions.mix(BUSCO.out.versions.first())
@@ -93,11 +94,12 @@ workflow BIN_QC {
         ch_bins_for_checkmlineagewf = ch_input_bins_for_qc
             .groupTuple()
             .filter { meta, _bins ->
-                    meta.domain != "eukarya"
-                }
+                meta.domain != "eukarya"
+            }
             .multiMap { meta, fa ->
                 reads: [meta, fa]
-                ext: fa.extension.unique().join("")  // the pipeline ensures that all bins will have the same extension
+                // the pipeline ensures that all bins will have the same extension
+                ext: fa.extension.unique().join("")  
             }
 
         CHECKM_LINEAGEWF(ch_bins_for_checkmlineagewf.reads, ch_bins_for_checkmlineagewf.ext, ch_checkm_db)
@@ -137,10 +139,11 @@ workflow BIN_QC {
                 meta.domain != "eukarya"
             }
             .flatMap { meta, bins ->
-                bins.collect { bin -> [meta, bin] }
+                // Set ID per bin, but save original ID for merging with CheckM output
+                bins.collect { bin -> [[id: bin.baseName, _id: meta.id] + meta, bin] }
             }
 
-        if ( params.gunc_db ) {
+        if (params.gunc_db) {
             ch_db_for_gunc = ch_gunc_db
         }
         else {
@@ -153,23 +156,32 @@ workflow BIN_QC {
 
         // Make sure to keep directory in sync with modules.conf
         GUNC_RUN.out.maxcss_level_tsv
-            .map{it[1]}
+            .map { it[1] }
             .collectFile(name: "gunc_summary.tsv", keepHeader: true, storeDir: "${params.outdir}/GenomeBinning/QC/")
 
-        if ( params.binqc_tool == 'checkm' ) {
-            ch_input_to_mergecheckm = GUNC_RUN.out.maxcss_level_tsv.combine(CHECKM_QA.out.output, by: 0)
+        if (params.binqc_tool == 'checkm') {
+            ch_input_to_gunc_combine = GUNC_RUN.out.maxcss_level_tsv
+                .map { meta, output ->
+                    // restore original ID to combine with CheckM output
+                    [[id: meta._id] + meta - meta.subMap(['_id']), output]
+                }
+                .groupTuple()
+
+            COMBINE_GUNC_TSV(ch_input_to_gunc_combine)
 
-            GUNC_MERGECHECKM(ch_input_to_mergecheckm)
+            ch_input_to_mergegunccheckm = COMBINE_BINQC_TSV.out.combined
+                .combine(CHECKM_QA.out.output, by: 0)
+
+            GUNC_MERGECHECKM(ch_input_to_mergegunccheckm)
             ch_versions.mix(GUNC_MERGECHECKM.out.versions)
 
             // Make sure to keep directory in sync with modules.conf
             GUNC_MERGECHECKM.out.tsv
-                .map{it[1]}
+                .map { it[1] }
                 .collectFile(name: "gunc_checkm_summary.tsv", keepHeader: true, storeDir: "${params.outdir}/GenomeBinning/QC/")
         }
     }
 
-
     emit:
     qc_summary      = qc_summary
     multiqc_files   = ch_multiqc_files