Run barrnap with fasta

CHANGELOG.md

@@ -18,6 +18,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - [#519](https://github.com/nf-core/ampliseq/pull/519) - Adding the pipeline reference to the MultiQC report
 - [#520](https://github.com/nf-core/ampliseq/pull/520),[#530](https://github.com/nf-core/ampliseq/pull/530) - Fix conda packages
 - [#531](https://github.com/nf-core/ampliseq/pull/531) - Update documentation
+- [#535](https://github.com/nf-core/ampliseq/pull/535) - Make sure barrnap runs with fasta input
 
 ### `Dependencies`
 

docs/usage.md

@@ -182,7 +182,8 @@ An [example samplesheet](../assets/samplesheet.tsv) has been provided with the p
 
 #### ASV/OTU fasta input
 
-When pointing at a file ending with `.fasta`, `.fna` or `.fa`, the containing ASV/OTU sequences will be taxonomically classified. All other pipeline steps will be skipped.
+When pointing at a file ending with `.fasta`, `.fna` or `.fa`, the containing ASV/OTU sequences will be taxonomically classified.
+Most of the steps of the pipeline will be skipped, but ITSx & Barrnap & length filtering can be applied before taxonomic classification.
 
 ```bash
 --input 'path/to/amplicon_sequences.fasta'

modules/local/filter_len_asv.nf

@@ -25,14 +25,16 @@ process FILTER_LEN_ASV {
     script:
     def min_len_asv = params.min_len_asv ?: '1'
     def max_len_asv = params.max_len_asv ?: '1000000'
+
+    def read_table  = table ? "table <- read.table(file = '$table', sep = '\t', comment.char = '', header=TRUE)" : "table <- data.frame(matrix(ncol = 1, nrow = 0))"
     """
     #!/usr/bin/env Rscript
 
     #load packages
     suppressPackageStartupMessages(library(Biostrings))
 
     #read abundance file, first column is ASV_ID
-    table <- read.table(file = "$table", sep = '\t', comment.char = "", header=TRUE)
+    $read_table
     colnames(table)[1] <- "ASV_ID"
 
     #read fasta file of ASV sequences

workflows/ampliseq.nf

@@ -201,7 +201,6 @@ workflow AMPLISEQ {
     //
     PARSE_INPUT ( params.input, is_fasta_input, single_end, params.multiple_sequencing_runs, params.extension )
     ch_reads = PARSE_INPUT.out.reads
-    ch_fasta = PARSE_INPUT.out.fasta
 
     //
     // MODULE: Rename files
@@ -305,29 +304,35 @@ workflow AMPLISEQ {
     // Modules : Filter rRNA
     // TODO: FILTER_SSU.out.stats needs to be merged still into "overall_summary.tsv"
     //
+    if ( is_fasta_input ) {
+        ch_unfiltered_fasta = PARSE_INPUT.out.fasta
+    } else {
+        ch_unfiltered_fasta = DADA2_MERGE.out.fasta
+    }
+
     if (!params.skip_barrnap && params.filter_ssu) {
-        BARRNAP ( DADA2_MERGE.out.fasta )
+        BARRNAP ( ch_unfiltered_fasta )
         ch_versions = ch_versions.mix(BARRNAP.out.versions.ifEmpty(null))
         FILTER_SSU ( DADA2_MERGE.out.fasta, DADA2_MERGE.out.asv, BARRNAP.out.matches )
         MERGE_STATS_FILTERSSU ( ch_stats, FILTER_SSU.out.stats )
         ch_stats = MERGE_STATS_FILTERSSU.out.tsv
         ch_dada2_fasta = FILTER_SSU.out.fasta
         ch_dada2_asv = FILTER_SSU.out.asv
     } else if (!params.skip_barrnap && !params.filter_ssu) {
-        BARRNAP ( DADA2_MERGE.out.fasta )
+        BARRNAP ( ch_unfiltered_fasta )
         ch_versions = ch_versions.mix(BARRNAP.out.versions.ifEmpty(null))
-        ch_dada2_fasta =  DADA2_MERGE.out.fasta
+        ch_dada2_fasta = ch_unfiltered_fasta
         ch_dada2_asv = DADA2_MERGE.out.asv
     } else {
-        ch_dada2_fasta =  DADA2_MERGE.out.fasta
+        ch_dada2_fasta = ch_unfiltered_fasta
         ch_dada2_asv = DADA2_MERGE.out.asv
     }
 
     //
     // Modules : amplicon length filtering
     //
     if (params.min_len_asv || params.max_len_asv) {
-        FILTER_LEN_ASV ( ch_dada2_fasta,ch_dada2_asv )
+        FILTER_LEN_ASV ( ch_dada2_fasta, ch_dada2_asv.ifEmpty( [] ) )
         ch_versions = ch_versions.mix(FILTER_LEN_ASV.out.versions.ifEmpty(null))
         MERGE_STATS_FILTERLENASV ( ch_stats, FILTER_LEN_ASV.out.stats )
         ch_stats = MERGE_STATS_FILTERLENASV.out.tsv
@@ -338,10 +343,7 @@ workflow AMPLISEQ {
     //
     // SUBWORKFLOW / MODULES : Taxonomic classification with DADA2 and/or QIIME2
     //
-    //Alternative entry point for fasta that is being classified
-    if ( !is_fasta_input ) {
-        ch_fasta = ch_dada2_fasta
-    }
+    ch_fasta = ch_dada2_fasta
 
     //DADA2
     if (!params.skip_taxonomy) {