Fix MaxBin2 abundance input

CHANGELOG.md

@@ -18,6 +18,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - [#674](https://github.com/nf-core/mag/pull/674) - Make longread preprocessing a subworkflow (added by @muabnezor)
 - [#674](https://github.com/nf-core/mag/pull/674) - Add porechop and filtlong logs to multiqc (added by @muabnezor)
 - [#674](https://github.com/nf-core/mag/pull/674) - Change local filtlong module to the official nf-core/filtlong module (added by @muabnezor)
+- [#690](https://github.com/nf-core/mag/pull/690) - MaxBin2 is using the abundance information from different samples now as expected (reported by @uel3 and fixed by @d4straub)
 
 ### `Dependencies`
 

conf/modules.config

@@ -657,6 +657,11 @@ process {
                 path: { "${params.outdir}/GenomeBinning/MaxBin2/discarded" },
                 mode: params.publish_dir_mode,
                 pattern: '*.tooshort.gz'
+            ],
+            [
+                path: { "${params.outdir}/GenomeBinning/MaxBin2/" },
+                mode: params.publish_dir_mode,
+                pattern: '*.{summary,abundance}'
             ]
         ]
         ext.prefix = { "${meta.assembler}-MaxBin2-${meta.id}" }

modules.json

@@ -169,7 +169,7 @@
                     },
                     "maxbin2": {
                         "branch": "master",
-                        "git_sha": "911696ea0b62df80e900ef244d7867d177971f73",
+                        "git_sha": "283613159e079152f1336cef0db1c836086206e0",
                         "installed_by": ["modules"]
                     },
                     "metabat2/jgisummarizebamcontigdepths": {

modules/local/convert_depths.nf

@@ -11,14 +11,27 @@ process CONVERT_DEPTHS {
 
     output:
     // need to add empty val because representing reads as we dont want maxbin to calculate for us.
-    tuple val(meta), path(fasta), val([]), path("*_mb2_depth.txt"), emit: output
-    path "versions.yml"                                           , emit: versions
+    tuple val(meta), path(fasta), val([]), path("*.abund"), emit: output
+    path "versions.yml"                                   , emit: versions
 
     script:
     def prefix = task.ext.prefix ?: "${meta.id}"
     """
     gunzip -f $depth
-    bioawk -t '{ { if (NR > 1) { { print \$1, \$3 } } } }' ${depth.toString() - '.gz'} > ${prefix}_mb2_depth.txt
+
+    # Determine the number of abundance columns
+    n_abund=\$(awk 'NR==1 {print int((NF-3)/2)}' ${depth.toString() - '.gz'})
+
+    # Get column names
+    read -r header<${depth.toString() - '.gz'}
+    header=(\$header)
+
+    # Generate abundance files for each read set
+    for i in \$(seq 1 \$n_abund); do
+        col=\$((i*2+2))
+        name=\$( echo \${header[\$col-1]} | sed s/\\.bam\$// )
+        bioawk -t '{if (NR > 1) {print \$1, \$'"\$col"'}}' ${depth.toString() - '.gz'} > \${name}.abund
+    done
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":