Clean up AdapterRemoval and remove all debugging dumps

subworkflows/local/db_check.nf

@@ -18,9 +18,7 @@ workflow DB_CHECK {
     parsed_samplesheet = DATABASE_CHECK ( dbsheet )
         .csv
         .splitCsv ( header:true, sep:',' )
-        .dump(tag: "db_split_csv_out")
         .map { create_db_channels(it) }
-        .dump(tag: "db_channel_prepped")
 
     ch_dbs_for_untar = parsed_samplesheet
         .branch {

subworkflows/local/input_check.nf

@@ -12,7 +12,6 @@ workflow INPUT_CHECK {
     parsed_samplesheet = SAMPLESHEET_CHECK ( samplesheet )
         .csv
         .splitCsv ( header:true, sep:',' )
-        .dump(tag: "input_split_csv_out")
         .branch {
             fasta: it['fasta'] != ''
             nanopore: it['instrument_platform'] == 'OXFORD_NANOPORE'
@@ -21,17 +20,14 @@ workflow INPUT_CHECK {
 
     parsed_samplesheet.fastq
         .map { create_fastq_channel(it) }
-        .dump(tag: "fastq_channel_init")
         .set { fastq }
 
     parsed_samplesheet.nanopore
         .map { create_fastq_channel(it) }
-        .dump(tag: "fastq_nanopore_channel_init")
         .set { nanopore }
 
     parsed_samplesheet.fasta
         .map { create_fasta_channel(it) }
-        .dump(tag: "fasta_channel_init")
         .set { fasta }
 
     emit:

subworkflows/local/longread_preprocessing.nf

@@ -16,7 +16,6 @@ workflow LONGREAD_PREPROCESSING {
     PORECHOP ( reads )
 
     ch_processed_reads = PORECHOP.out.reads
-                                .dump(tag: "pre_fastqc_check")
                                 .map {
                                         meta, reads ->
                                         def meta_new = meta.clone()

subworkflows/local/shortread_adapterremoval.nf

@@ -24,9 +24,10 @@ workflow SHORTREAD_ADAPTERREMOVAL {
     ADAPTERREMOVAL_SINGLE ( ch_input_for_adapterremoval.single, [] )
     ADAPTERREMOVAL_PAIRED ( ch_input_for_adapterremoval.paired, [] )
 
-    if ( params.shortread_clipmerge_mergepairs && !params.shortread_clipmerge_excludeunmerged ) {
+    // due to the slightly ugly output implementation of the current AdapterRemoval2 version, each file
-        // due to the slightly ugly output implementation of the current AdapterRemoval2 module, each file
     // has to be exported in a separate channel, and we must manually recombine when necessary
+
+    if ( params.shortread_clipmerge_mergepairs && !params.shortread_clipmerge_excludeunmerged ) {
         ch_adapterremoval_for_cat = ADAPTERREMOVAL_PAIRED.out.collapsed
                                                 .mix(
                                                     ADAPTERREMOVAL_PAIRED.out.collapsed_truncated,
@@ -43,8 +44,8 @@ workflow SHORTREAD_ADAPTERREMOVAL {
                                                     }
                                                     .groupTuple()
 
-
+        ch_adapterremoval_reads_prepped = CAT_FASTQ ( ch_adapterremoval_for_cat ).reads
-        ch_adapterremoval_reads_prepped = CAT_FASTQ ( ch_adapterremoval_for_cat ).reads.mix( ADAPTERREMOVAL_SINGLE.out.singles_truncated )
+                                            .mix( ADAPTERREMOVAL_SINGLE.out.singles_truncated )
 
     } else if ( params.shortread_clipmerge_mergepairs && params.shortread_clipmerge_excludeunmerged ) {
         ch_adapterremoval_for_cat = ADAPTERREMOVAL_PAIRED.out.collapsed
@@ -58,7 +59,8 @@ workflow SHORTREAD_ADAPTERREMOVAL {
                                                     }
                                                     .groupTuple(by: 0)
 
-            ch_adapterremoval_reads_prepped = CAT_FASTQ ( ch_adapterremoval_for_cat ).reads.mix( ADAPTERREMOVAL_SINGLE.out.singles_truncated )
+        ch_adapterremoval_reads_prepped = CAT_FASTQ ( ch_adapterremoval_for_cat ).reads
+                                            .mix( ADAPTERREMOVAL_SINGLE.out.singles_truncated )
 
     } else {
 

workflows/taxprofiler.nf

@@ -18,6 +18,7 @@ for (param in checkPathParamList) { if (param) { file(param, checkIfExists: true
 if (params.input    ) { ch_input     = file(params.input)     } else { exit 1, 'Input samplesheet not specified!' }
 if (params.databases) { ch_databases = file(params.databases) } else { exit 1, 'Input database sheet not specified!' }
 if (params.shortread_clipmerge_mergepairs && params.run_malt ) log.warn "[nf-core/taxprofiler] warning: MALT does not except uncollapsed paired-reads. Pairs will be profiled as separate files."
+if (params.shortread_clipmerge_excludeunmerged && !params.shortread_clipmerge_mergepairs) exit 1, "[nf-core/taxprofiler] error: cannot include unmerged reads when merging not turned on. Please specify --shortread_clipmerge_mergepairs"
 
 /*
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@@ -126,7 +127,6 @@ workflow TAXPROFILER {
     ch_input_for_profiling = ch_shortreads_preprocessed
             .mix( ch_longreads_preprocessed )
             .combine(DB_CHECK.out.dbs)
-            .dump(tag: "reads_plus_db_clean")
             .branch {
                 malt:    it[2]['tool'] == 'malt'
                 kraken2: it[2]['tool'] == 'kraken2'
@@ -141,9 +141,7 @@ workflow TAXPROFILER {
     // loading takes a long time, so we only want to run it once per database
     // TODO document somewhere we only accept illumina short reads for MALT?
     ch_input_for_malt =  ch_input_for_profiling.malt
-                            .dump(tag: "input_to_malt_prefilter")
                             .filter { it[0]['instrument_platform'] == 'ILLUMINA' }
-                            .dump(tag: "input_to_malt_postfilter")
                             .map {
                                 it ->
                                     def temp_meta =  [ id: it[2]['db_name']]  + it[2]
@@ -151,7 +149,6 @@ workflow TAXPROFILER {
                                     [ temp_meta, it[1], db ]
                             }
                             .groupTuple(by: [0,2])
-                            .dump(tag: "input_to_malt")
                             .multiMap {
                                 it ->
                                     reads: [ it[0], it[1].flatten() ]
@@ -160,7 +157,6 @@ workflow TAXPROFILER {
 
     // We can run Kraken2 one-by-one sample-wise
     ch_input_for_kraken2 =  ch_input_for_profiling.kraken2
-                            .dump(tag: "input_to_kraken")
                             .multiMap {
                                 it ->
                                     reads: [ it[0] + it[2], it[1] ]