fix: Make CheckM auto-detect file extensions (#76)

* test: Compress test data FASTA files * test: Change tests to use Channel.of * Revert "test: Compress test data FASTA files" This reverts commit 673fc7e. * test: Remove duplicated assignment * refactor: Move assemblyqc subwf into separate wf file * fix: Change imports * feat: Auto-detect fasta file extension * test: Add assemblyqc test * test: Change paths to file values * test: Remove vibrant from testing - Because its db fills up the runner storage * refactor: Remove meta.id from checkm outdir * Revert "refactor: Remove meta.id from checkm outdir" This reverts commit ec98f36. * test: Skip assemblyqc test on the CI * tests: Fix trace size * chore: Bump version and changelog
beiko-lab · Apr 10, 2023 · 4f20672 · 4f20672
1 parent aef2f6c
commit 4f20672
Show file tree

Hide file tree

Showing 7 changed files with 123 additions and 43 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -3,6 +3,17 @@
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/)
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## v2.0.1 dev - April 10, 2023
+
+### `Fixed`
+
+- Makes CheckM auto-detect FASTA file extensions
+  - So it doesn't fail with files without `.fna`
+
+### `Added`
+
+- Moves Assembly QC subworkflow into its own file
+
 ## v2.0 dev - April 5, 2023
 
 ### `Added`

diff --git a/nextflow.config b/nextflow.config
@@ -180,7 +180,7 @@ manifest {
     description     = 'AMR/VF LGT-focused bacterial genomics workflow'
     mainScript      = 'main.nf'
     nextflowVersion = '!>=21.04.0'
-    version = '2.0dev'
+    version = '2.0.1dev'
 }
 
 // Function to ensure that resource requirements don't go beyond

diff --git a/subworkflows/local/assemblyqc.nf b/subworkflows/local/assemblyqc.nf
@@ -0,0 +1,60 @@
+include { GET_DB_CACHE } from '../../modules/local/get_db_cache'
+include { QUAST } from '../../modules/nf-core/quast/main'
+include { KRAKEN2_KRAKEN2 as KRAKEN2_RUN } from '../../modules/nf-core/kraken2/kraken2/main'
+include { CHECKM_LINEAGEWF } from '../../modules/nf-core/checkm/lineagewf/main'
+
+workflow CHECK_ASSEMBLIES {
+    take:
+        assemblies
+        krakendb_cache
+        reference_genome
+        use_reference_genome
+
+    main:
+
+        ch_multiqc_files = Channel.empty()
+        ch_software_versions = Channel.empty()
+
+        ///*
+        // * MODULE: Run Kraken2
+        // */
+        if (!params.skip_kraken) {
+            if(krakendb_cache) {
+                GET_DB_CACHE(krakendb_cache)
+                KRAKEN2_RUN(assemblies, GET_DB_CACHE.out.minikraken, false, true)
+            } else {
+                KRAKEN2_DB()
+                KRAKEN2_RUN(assemblies, KRAKEN2_DB.out.minikraken, false, true)
+            }
+
+            ch_software_versions = ch_software_versions.mix(KRAKEN2_RUN.out.versions.first().ifEmpty(null))
+            ch_multiqc_files = ch_multiqc_files.mix(KRAKEN2_RUN.out.report.collect{it[1]}.ifEmpty([]))
+        }
+
+        fasta_extension = assemblies.map{ id, path -> path.getExtension() }.first()
+
+        /*
+        * Module: CheckM Quality Check
+        */
+        CHECKM_LINEAGEWF(assemblies, fasta_extension, [])
+        ch_software_versions = ch_software_versions.mix(CHECKM_LINEAGEWF.out.versions.first().ifEmpty(null))
+        /*
+        * Module: QUAST quality check
+        */
+        // Need to reformat assembly channel for QUAST
+        // pattern adapted from nf-core/bacass
+        ch_assembly = Channel.empty()
+        ch_assembly = ch_assembly.mix(assemblies.dump(tag: 'assembly'))
+        ch_assembly
+            .map { meta, fasta -> fasta } //QUAST doesn't take the meta tag
+            .collect()
+            .set { ch_to_quast }
+        QUAST(ch_to_quast, reference_genome, [], use_reference_genome, false)
+        ch_software_versions = ch_software_versions.mix(QUAST.out.versions.ifEmpty(null))
+
+        ch_multiqc_files = ch_multiqc_files.mix(QUAST.out.tsv.collect())
+
+    emit:
+        assemblyqc_software = ch_software_versions
+        multiqc = ch_multiqc_files
+}
diff --git a/tests/subworkflows/local/assemblyqc.nf.test.skip b/tests/subworkflows/local/assemblyqc.nf.test.skip
@@ -0,0 +1,38 @@
+nextflow_workflow {
+
+    name "Test Workflow CHECK_ASSEMBLIES"
+    script "subworkflows/local/assemblyqc.nf"
+    workflow "CHECK_ASSEMBLIES"
+
+    test("Assembly QC runs without failures") {
+
+        when {
+            params {
+                skip_kraken = true
+                outdir = "$outputDir"
+            }
+            workflow {
+                """
+                // define inputs of the workflow here. Example:
+                input[0] = Channel.of(
+                    [[id:'SRR14022735'], file("$baseDir/test/SRR14022735_T1.scaffolds.fa")],
+                    [[id:'SRR14022737'], file("$baseDir/test/SRR14022737_T1.scaffolds.fa")],
+                    [[id:'SRR14022754'], file("$baseDir/test/SRR14022754_T1.scaffolds.fa")],
+                    [[id:'SRR14022764'], file("$baseDir/test/SRR14022764_T1.scaffolds.fa")],
+                )
+                input[1] = []
+                // No ref genome
+                input[2] = []
+                input[3] = false
+                """
+            }
+        }
+
+        then {
+            assert workflow.success
+            assert workflow.trace.tasks().size() >= 2
+        }
+
+    }
+
+}
diff --git a/tests/subworkflows/local/phylo.nf.test b/tests/subworkflows/local/phylo.nf.test
@@ -14,11 +14,11 @@ nextflow_workflow {
             workflow {
                 """
                 // define inputs of the workflow here. Example:
-                input[0] = ch = Channel.from([
+                input[0] = Channel.of(
                     [[id:'SRR14022735'], '$baseDir/test/SRR14022735_T1.gff'],
                     [[id:'SRR14022764'], '$baseDir/test/SRR14022764_T1.gff'],
                     [[id:'SRR14022754'], '$baseDir/test/SRR14022754_T1.gff']
-                ])
+                )
                 // Don't use full alignment
                 input[1] = false
                 // Use fasttree

diff --git a/tests/subworkflows/local/poppunk.nf.test b/tests/subworkflows/local/poppunk.nf.test
@@ -15,12 +15,12 @@ nextflow_workflow {
             workflow {
                 """
                 // define inputs of the workflow here. Example:
-                input[0] = Channel.from([
+                input[0] = Channel.of(
                     [[id:'SRR14022735'], "$baseDir/test/SRR14022735_T1.scaffolds.fa"],
                     [[id:'SRR14022737'], "$baseDir/test/SRR14022737_T1.scaffolds.fa"],
                     [[id:'SRR14022754'], "$baseDir/test/SRR14022754_T1.scaffolds.fa"],
                     [[id:'SRR14022764'], "$baseDir/test/SRR14022764_T1.scaffolds.fa"],
-                ])
+                )
                 """
             }
         }

diff --git a/workflows/arete.nf b/workflows/arete.nf
@@ -42,6 +42,7 @@ include { INPUT_CHECK;
 
 include { ASSEMBLE_SHORTREADS } from '../subworkflows/local/assembly'
 include { ANNOTATE_ASSEMBLIES } from '../subworkflows/local/annotation'
+include { CHECK_ASSEMBLIES } from '../subworkflows/local/assemblyqc'
 include { PHYLOGENOMICS } from '../subworkflows/local/phylo'
 include { RUN_POPPUNK } from '../subworkflows/local/poppunk'
 include { SUBSET_GENOMES } from '../subworkflows/local/subsample'
@@ -59,8 +60,6 @@ include { MULTIQC } from '../modules/nf-core/multiqc/main'
 include { FASTQC as TRIM_FASTQC } from '../modules/nf-core/fastqc/main'
 include { FASTP                 } from '../modules/nf-core/fastp/main'
 include { UNICYCLER             } from '../modules/nf-core/unicycler/main'
-include { QUAST                 } from '../modules/nf-core/quast/main'
-include { KRAKEN2_KRAKEN2 as KRAKEN2_RUN } from '../modules/nf-core/kraken2/kraken2/main'
 include { PROKKA                } from '../modules/nf-core/prokka/main'
 include { GET_CAZYDB;
           GET_VFDB;
@@ -73,7 +72,6 @@ include { DIAMOND_BLASTX as DIAMOND_BLAST_CAZY;
           DIAMOND_BLASTX as DIAMOND_BLAST_BACMET } from '../modules/nf-core/diamond/blastx/main'
 include { IQTREE } from '../modules/nf-core/iqtree/main'
 include { SNPSITES } from '../modules/nf-core/snpsites/main'
-include { CHECKM_LINEAGEWF } from '../modules/nf-core/checkm/lineagewf/main'
 //
 // MODULE: Local to the pipeline
 //
@@ -446,41 +444,14 @@ workflow QUALITYCHECK{
      */
     ANNOTATION_INPUT_CHECK(ch_input)
 
-    ///*
-    // * MODULE: Run Kraken2
-    // */
-    if (!params.skip_kraken) {
-        if(db_cache) {
-            GET_DB_CACHE(db_cache)
-            KRAKEN2_RUN(ANNOTATION_INPUT_CHECK.out.genomes, GET_DB_CACHE.out.minikraken, false, true)
-        } else {
-            KRAKEN2_DB()
-            KRAKEN2_RUN(ANNOTATION_INPUT_CHECK.out.genomes, KRAKEN2_DB.out.minikraken, false, true)
-        }
-
-        ch_software_versions = ch_software_versions.mix(KRAKEN2_RUN.out.versions.first().ifEmpty(null))
-        ch_multiqc_files = ch_multiqc_files.mix(KRAKEN2_RUN.out.report.collect{it[1]}.ifEmpty([]))
-    }
-    /*
-    * Module: CheckM Quality Check
-    */
-    CHECKM_LINEAGEWF(ANNOTATION_INPUT_CHECK.out.genomes, "fna", []) //todo figure out a way to infer the file extension during input check
-    ch_software_versions = ch_software_versions.mix(CHECKM_LINEAGEWF.out.versions.first().ifEmpty(null))
-    /*
-     * Module: QUAST quality check
-     */
-    // Need to reformat assembly channel for QUAST
-    // pattern adapted from nf-core/bacass
-    ch_assembly = Channel.empty()
-    ch_assembly = ch_assembly.mix(ANNOTATION_INPUT_CHECK.out.genomes.dump(tag: 'assembly'))
-    ch_assembly
-        .map { meta, fasta -> fasta } //QUAST doesn't take the meta tag
-        .collect()
-        .set { ch_to_quast }
-    QUAST(ch_to_quast, ch_reference_genome, [], use_reference_genome, false)
-    ch_software_versions = ch_software_versions.mix(QUAST.out.versions.ifEmpty(null))
-
-    ch_multiqc_files = ch_multiqc_files.mix(QUAST.out.tsv.collect())
+    CHECK_ASSEMBLIES(
+        ANNOTATION_INPUT_CHECK.out.genomes,
+        db_cache,
+        ch_reference_genome,
+        use_reference_genome
+    )
+    ch_software_versions = ch_software_versions.mix(CHECK_ASSEMBLIES.out.assemblyqc_software)
+    ch_multiqc_files = ch_multiqc_files.mix(CHECK_ASSEMBLIES.out.multiqc)
 
     MULTIQC(
         ch_multiqc_files.collect(),