Merge pull request #258 from nf-core/clone-maps

refactor: double check maps and validation

subworkflows/local/db_check.nf

@@ -28,22 +28,20 @@ workflow DB_CHECK {
     // Normal checks for within-row validity, so can be moved to separate functions
     parsed_samplesheet = Channel.fromPath(dbsheet)
         .splitCsv ( header:true, sep:',' )
-        .map {
+        .map { row ->
-            validate_db_rows(it)
+            validate_db_rows(row)
-            create_db_channels(it)
+            return [ row.subMap(['tool', 'db_name', 'db_params']), file(row.db_path) ]
         }
 
     ch_dbs_for_untar = parsed_samplesheet
-        .branch {
+        .branch { db_meta, db ->
-            untar: it[1].toString().endsWith(".tar.gz")
+            untar: db.name.endsWith(".tar.gz")
             skip: true
         }
 
     // Filter the channel to untar only those databases for tools that are selected to be run by the user.
     ch_input_untar = ch_dbs_for_untar.untar
-                        .filter {
+        .filter { db_meta, db -> params["run_${db_meta.tool}"] }
-                          params["run_${it[0]['tool']}"]
-                        }
 
     UNTAR (ch_input_untar)
     ch_versions = ch_versions.mix(UNTAR.out.versions.first())
@@ -61,7 +59,7 @@ def validate_db_rows(LinkedHashMap row){
 
     // all columns there
     def expected_headers = ['tool', 'db_name', 'db_params', 'db_path']
-        if ( !row.keySet().containsAll(expected_headers) ) exit 1, "[nf-core/taxprofiler] ERROR: Invalid database input sheet - malformed column names. Please check input TSV. Column names should be: ${expected_keys.join(", ")}"
+    if ( !row.keySet().containsAll(expected_headers) ) exit 1, "[nf-core/taxprofiler] ERROR: Invalid database input sheet - malformed column names. Please check input TSV. Column names should be: ${expected_headers.join(", ")}"
 
     // valid tools specified
     def expected_tools = [ "bracken", "centrifuge", "diamond", "kaiju", "kraken2", "krakenuniq", "malt", "metaphlan3", "motus" ]
@@ -74,21 +72,7 @@ def validate_db_rows(LinkedHashMap row){
     // check if any form of bracken params, that it must have `;`
     if ( row.tool == 'bracken' && row.db_params && !row.db_params.contains(";") )  exit 1, "[nf-core/taxprofiler] ERROR: Invalid database db_params entry. Bracken requires a semi-colon if passing parameter. Error in: ${row}"
 
+    // ensure that the database directory exists
+    if (!file(row.db_path, type: 'dir').exists()) exit 1, "ERROR: Please check input samplesheet -> database path could not be found!\n${row.db_path}"
+
 }
-
-def create_db_channels(LinkedHashMap row) {
-    def meta = [:]
-    meta.tool             = row.tool
-    meta.db_name          = row.db_name
-    meta.db_params        = row.db_params
-
-    def array = []
-    if (!file(row.db_path, type: 'dir').exists()) {
-        exit 1, "ERROR: Please check input samplesheet -> database path could not be found!\n${row.db_path}"
-    }
-    array = [ meta, file(row.db_path) ]
-
-    return array
-}
-
-

subworkflows/local/input_check.nf

@@ -12,9 +12,9 @@ workflow INPUT_CHECK {
     parsed_samplesheet = SAMPLESHEET_CHECK ( samplesheet )
         .csv
         .splitCsv ( header:true, sep:',' )
-        .branch {
+        .branch { row ->
-            fasta: it['fasta'] != ''
+            fasta: row.fasta != ''
-            nanopore: it['instrument_platform'] == 'OXFORD_NANOPORE'
+            nanopore: row.instrument_platform == 'OXFORD_NANOPORE'
             fastq: true
         }
 
@@ -37,49 +37,42 @@ workflow INPUT_CHECK {
 // Function to get list of [ meta, [ fastq_1, fastq_2 ] ]
 def create_fastq_channel(LinkedHashMap row) {
     // create meta map
-    def meta = [:]
+    def meta = row.subMap(['sample', 'run_accession', 'instrument_platform'])
-    meta.id                     = row.sample
+    meta.id         = meta.sample
-    meta.run_accession          = row.run_accession
-    meta.instrument_platform    = row.instrument_platform
     meta.single_end = row.single_end.toBoolean()
     meta.is_fasta   = false
 
     // add path(s) of the fastq file(s) to the meta map
-    def fastq_meta = []
     if (!file(row.fastq_1).exists()) {
         exit 1, "ERROR: Please check input samplesheet -> Read 1 FastQ file does not exist!\n${row.fastq_1}"
     }
+
     if (meta.single_end) {
-        fastq_meta = [ meta, [ file(row.fastq_1) ] ]
+        return [ meta, [ file(row.fastq_1) ] ]
     } else {
         if (meta.instrument_platform == 'OXFORD_NANOPORE') {
             if (row.fastq_2 != '') {
                 exit 1, "ERROR: Please check input samplesheet -> For Oxford Nanopore reads Read 2 FastQ should be empty!\n${row.fastq_2}"
             }
-            fastq_meta = [ meta, [ file(row.fastq_1) ] ]
+            return [ meta, [ file(row.fastq_1) ] ]
         } else {
             if (!file(row.fastq_2).exists()) {
                 exit 1, "ERROR: Please check input samplesheet -> Read 2 FastQ file does not exist!\n${row.fastq_2}"
             }
-         fastq_meta = [ meta, [ file(row.fastq_1), file(row.fastq_2) ] ]
+            return [ meta, [ file(row.fastq_1), file(row.fastq_2) ] ]
+        }
+    }
 }
 
-    }
+// Function to get list of [ meta, fasta ]
-    return fastq_meta
-}// Function to get list of [ meta, fasta ]
 def create_fasta_channel(LinkedHashMap row) {
-    def meta = [:]
+    def meta        = row.subMap(['sample', 'run_accession', 'instrument_platform'])
-    meta.id                     = row.sample
+    meta.id         = meta.sample
-    meta.run_accession          = row.run_accession
-    meta.instrument_platform    = row.instrument_platform
     meta.single_end = true
     meta.is_fasta   = true
 
-    def array = []
     if (!file(row.fasta).exists()) {
         exit 1, "ERROR: Please check input samplesheet -> FastA file does not exist!\n${row.fasta}"
     }
-    array = [ meta, [ file(row.fasta) ] ]
+    return [ meta, [ file(row.fasta) ] ]
-
-    return array
 }

subworkflows/local/longread_hostremoval.nf

@@ -46,7 +46,7 @@ workflow LONGREAD_HOSTREMOVAL {
     ch_versions      = ch_versions.mix( SAMTOOLS_INDEX.out.versions.first() )
 
     bam_bai = MINIMAP2_ALIGN.out.bam
-        .join(SAMTOOLS_INDEX.out.bai, remainder: true)
+        .join(SAMTOOLS_INDEX.out.bai)
 
     SAMTOOLS_STATS ( bam_bai, reference )
     ch_versions = ch_versions.mix(SAMTOOLS_STATS.out.versions.first())

subworkflows/local/longread_preprocessing.nf

@@ -20,12 +20,7 @@ workflow LONGREAD_PREPROCESSING {
         PORECHOP_PORECHOP ( reads )
 
         ch_processed_reads = PORECHOP_PORECHOP.out.reads
-                                        .map {
+            .map { meta, reads -> [ meta + [single_end: 1], reads ] }
-                                                meta, reads ->
-                                                def meta_new = meta.clone()
-                                                meta_new['single_end'] = 1
-                                                [ meta_new, reads ]
-                                        }
 
         ch_versions = ch_versions.mix(PORECHOP_PORECHOP.out.versions.first())
         ch_multiqc_files = ch_multiqc_files.mix( PORECHOP_PORECHOP.out.log )
@@ -39,12 +34,7 @@ workflow LONGREAD_PREPROCESSING {
     } else {
         PORECHOP_PORECHOP ( reads )
         ch_clipped_reads = PORECHOP_PORECHOP.out.reads
-                                        .map {
+            .map { meta, reads -> [ meta + [single_end: 1], reads ] }
-                                                meta, reads ->
-                                                def meta_new = meta.clone()
-                                                meta_new['single_end'] = 1
-                                                [ meta_new, reads ]
-                                        }
 
         ch_processed_reads = FILTLONG ( ch_clipped_reads.map { meta, reads -> [ meta, [], reads ] } ).reads
 

subworkflows/local/profiling.nf

@@ -35,10 +35,7 @@ workflow PROFILING {
     ch_input_for_profiling = reads
             .map {
                 meta, reads ->
-                    def meta_new = meta.clone()
+                    [meta + [id: "${meta.id}${meta.single_end ? '_se' : '_pe'}"], reads]
-                        pairtype = meta_new['single_end'] ? '_se' : '_pe'
-                        meta_new['id'] =  meta_new['id'] + pairtype
-                    [meta_new, reads]
             }
             .combine(databases)
             .branch {
@@ -92,9 +89,9 @@ workflow PROFILING {
             }
             .groupTuple(by: [0,2])
             .multiMap {
-                                    it ->
+                meta, reads, db ->
-                                        reads: [ it[0], it[1].flatten() ]
+                    reads: [ meta, reads.flatten() ]
-                                        db: it[2]
+                    db: db
             }
 
         MALT_RUN ( ch_input_for_malt.reads, ch_input_for_malt.db )