Added nftests to gseagsea (#7266)

* Added nftests to gseagsea

* Update input files

* fix nftests

* Address PR comments

* Add seed setting parameter for stabilizing results

* Fix linting

* Add default random seed

* Update datasets paths

---------

Co-authored-by: Simon Pearce <[email protected]>

modules/nf-core/gsea/gsea/main.nf

@@ -4,8 +4,8 @@ process GSEA_GSEA {
 
     conda "${moduleDir}/environment.yml"
     container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
-        'https://depot.galaxyproject.org/singularity/gsea:4.3.2--hdfd78af_0':
-        'biocontainers/gsea:4.3.2--hdfd78af_0' }"
+        'https://community-cr-prod.seqera.io/docker/registry/v2/blobs/sha256/0f/0f4fe28961396eeeaa98484cb4f2db5c79abfdf117700df132312fe5c41bff81/data':
+        'community.wave.seqera.io/library/gsea:4.3.2--a7421d7504fd7c81' }"
 
     input:
     tuple val(meta), path(gct), path(cls), path(gene_sets)
@@ -26,29 +26,27 @@ process GSEA_GSEA {
     tuple val(meta), path("*heat_map_1.png")                   , emit: heatmap
     tuple val(meta), path("*pvalues_vs_nes_plot.png")          , emit: pvalues_vs_nes_plot
     tuple val(meta), path("*ranked_list_corr_2.png")           , emit: ranked_list_corr
-    tuple val(meta), path("*butterfly_plot.png")               , emit: butterfly_plot, optional: true
-    tuple val(meta), path("*[!gene_set_size|gsea_report|ranked_gene_list]*.tsv"), emit: gene_set_tsv, optional: true
-    tuple val(meta), path("*[!gsea_report|heat_map_corr_plot|index|pos_snapshot|neg_snapshot]*.html"), emit: gene_set_html, optional: true
-    tuple val(meta), path("*[!butterfly|enplot|global_es_histogram|gset_rnd_es_dist|heat_map|pvalues_vs_nes_plot|ranked_list_corr]*.png"), emit: gene_set_heatmap, optional: true
-    tuple val(meta), path("*_snapshot*.html")                  , emit: snapshot, optional: true
-    tuple val(meta), path("*enplot*.png")                      , emit: gene_set_enplot, optional: true
-    tuple val(meta), path("*gset_rnd_es_dist*.png")            , emit: gene_set_dist, optional: true
-    tuple val(meta), path("*.zip")                             , emit: archive, optional: true
+    tuple val(meta), path("*butterfly_plot.png")               , emit: butterfly_plot  , optional: true
+    tuple val(meta), path("gene_sets_*.tsv")                   , emit: gene_set_tsv    , optional: true
+    tuple val(meta), path("gene_sets_*.html")                  , emit: gene_set_html   , optional: true
+    tuple val(meta), path("gene_sets_*.png")                   , emit: gene_set_heatmap, optional: true
+    tuple val(meta), path("*_snapshot*.html")                  , emit: snapshot        , optional: true
+    tuple val(meta), path("*enplot*.png")                      , emit: gene_set_enplot , optional: true
+    tuple val(meta), path("*gset_rnd_es_dist*.png")            , emit: gene_set_dist   , optional: true
+    tuple val(meta), path("*.zip")                             , emit: archive         , optional: true
     path "versions.yml"                                        , emit: versions
 
     when:
     task.ext.when == null || task.ext.when
 
     script:
-    def VERSION = '4.3.2' // WARN: Version information not provided by tool on CLI. Please update this string when bumping container versions.
     def args = task.ext.args ?: ''
     def prefix = task.ext.prefix ?: "${meta.id}"
-
-    // Remove any trailing dots from prefix when passed as report label, so
-    // GSEA doesn't produce double-dotted top-level outputs
-    def rpt_label = prefix.replaceAll('\\.$', '')
-
+    def rpt_label = prefix.replaceAll('\\.$', '') // Remove any trailing dots from prefix when passed as report label, so GSEA doesn't produce double-dotted top-level outputs
     def chip_command = chip ? "-chip $chip -collapse true" : ''
+    def VERSION = '4.3.2' // WARN: Version information not provided by tool on CLI. Please update this string when bumping container versions.
+    if (!(args ==~ /.*-rnd_seed.*/)) {args += " -rnd_seed 10"}
+
     """
     # Run GSEA
 
@@ -78,9 +76,71 @@ process GSEA_GSEA {
         sed -i.bak "s/\$f/${prefix}\${f}/g" *.rpt *.html && rm *.bak
     done
 
+    # Rename files so that they can be properly referenced by the output channels
+    # Function to rename files based on the given pattern
+    rename_files() {
+        local pattern=\$1
+        local exclude_patterns=\$2
+        local extension=\$3
+
+        # Find files matching the pattern but not matching the exclusion patterns
+        find . -type f -name "\$pattern" | while read -r file; do
+            # Exclude files based on the provided exclusion patterns
+            if ! echo "\$file" | grep -qE "\$exclude_patterns"; then
+                # Rename the file by adding the prefix "gene_sets_"
+                mv "\$file" "\$(dirname "\$file")/gene_sets_\$(basename "\$file")"
+            fi
+        done
+    }
+
+    # Pattern and exclusion for .tsv files
+    tsv_pattern="*.tsv"
+    tsv_exclude="gene_set_size|gsea_report|ranked_gene_list"
+
+    # Pattern and exclusion for .html files
+    html_pattern="*.html"
+    html_exclude="gsea_report|heat_map_corr_plot|index|pos_snapshot|neg_snapshot"
+
+    # Pattern and exclusion for .png files
+    png_pattern="*.png"
+    png_exclude="butterfly|enplot|global_es_histogram|gset_rnd_es_dist|heat_map|pvalues_vs_nes_plot|ranked_list_corr"
+
+    # Rename .tsv files
+    rename_files "\$tsv_pattern" "\$tsv_exclude" ".tsv"
+
+    # Rename .html files
+    rename_files "\$html_pattern" "\$html_exclude" ".html"
+
+    # Rename .png files
+    rename_files "\$png_pattern" "\$png_exclude" ".png"
+
+
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":
         gsea: $VERSION
     END_VERSIONS
     """
+
+    stub:
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    def VERSION = '4.3.2' // WARN: Version information not provided by tool on CLI. Please update this string when bumping container versions.
+    """
+    touch ${prefix}.rpt
+    touch ${prefix}.index.html
+    touch ${prefix}.heat_map_corr_plot.html
+    touch ${prefix}.gsea_report_for_${reference}.tsv
+    touch ${prefix}.gsea_report_for_${reference}.html
+    touch ${prefix}.gsea_report_for_${target}.tsv
+    touch ${prefix}.gsea_report_for_${target}.html
+    touch ${prefix}.ranked_gene_list*.tsv
+    touch ${prefix}.gene_set_sizes.tsv
+    touch ${prefix}.global_es_histogram.png
+    touch ${prefix}.heat_map_1.png
+    touch ${prefix}.pvalues_vs_nes_plot.png
+    touch ${prefix}.ranked_list_corr_2.png
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        gsea: $VERSION
+    """
 }

modules/nf-core/gsea/gsea/meta.yml

@@ -182,31 +182,31 @@ output:
           type: map
           description: |
             Groovy Map containing metadata, at a minimum an id e.g. [ id:'test' ]
-      - "*[!gene_set_size|gsea_report|ranked_gene_list]*.tsv":
+      - "gene_sets_*.tsv":
           type: list
           description: Where -make_sets is not set to false, TSV files, one file for each
             gene set, with detail on enrichment for each gene
-          pattern: "[!gene_set_size|gsea_report|ranked_gene_list]*.tsv"
+          pattern: "gene_sets_*.tsv"
   - gene_set_html:
       - meta:
           type: map
           description: |
             Groovy Map containing metadata, at a minimum an id e.g. [ id:'test' ]
-      - "*[!gsea_report|heat_map_corr_plot|index|pos_snapshot|neg_snapshot]*.html":
+      - "gene_sets_*.html":
           type: list
           description: Where -make_sets is not set to false, HTML files, one file for
             each gene set, with detail on enrichment for each gene
-          pattern: "[!gsea_report|heat_map_corr_plot|index|pos_snapshot|neg_snapshot]*.html"
+          pattern: "gene_sets_*.html"
   - gene_set_heatmap:
       - meta:
           type: map
           description: |
             Groovy Map containing metadata, at a minimum an id e.g. [ id:'test' ]
-      - "*[!butterfly|enplot|global_es_histogram|gset_rnd_es_dist|heat_map|pvalues_vs_nes_plot|ranked_list_corr]*.png":
+      - "gene_sets_*.png":
           type: list
           description: Where -make_sets is not set to false, PNG-format heatmaps, one
             file for each gene set, showing expression for each gene
-          pattern: "[!butterfly|enplot|global_es_histogram|gset_rnd_es_dist|heat_map|pvalues_vs_nes_plot|ranked_list_corr]*.png"
+          pattern: "gene_sets_*.png"
   - snapshot:
       - meta:
           type: map
@@ -255,3 +255,4 @@ authors:
   - "@pinin4fjords"
 maintainers:
   - "@pinin4fjords"
+  - "@nschcolnicov"

modules/nf-core/gsea/gsea/tests/main.nf.test

@@ -0,0 +1,78 @@
+nextflow_process {
+
+    name "Test Process GSEA_GSEA"
+    script "../main.nf"
+    process "GSEA_GSEA"
+
+    tag "modules"
+    tag "modules_nfcore"
+    tag "gsea_gsea"
+    tag "gsea"
+    tag "gsea/gsea"
+
+    test("test") {
+
+        when {
+            process {
+                """
+                input[0] = [['id':'Condition_genotype_WT_KO', 'variable':'Condition genotype', 'reference':'WT', 'target':'KO', 'blocking':'batch'], file("https://github.com/nf-core/test-datasets/raw/refs/heads/modules/data/genomics/mus_musculus/gene_set_analysis/Condition_treatment_Control_Treated.gct", checkIfExists:true), file("https://github.com/nf-core/test-datasets/raw/refs/heads/modules/data/genomics/mus_musculus/gene_set_analysis/Condition_genotype_WT_KO.cls", checkIfExists:true), file("https://github.com/nf-core/test-datasets/raw/refs/heads/modules/data/genomics/mus_musculus/gene_set_analysis/mh.all.v2022.1.Mm.symbols.gmt", checkIfExists:true)]
+                input[1] = ['WT', 'KO']
+                input[2] = file("https://github.com/nf-core/test-datasets/raw/refs/heads/modules/data/genomics/mus_musculus/gene_set_analysis/Mus_musculus.anno.feature_metadata.chip", checkIfExists:true)
+                """
+            }
+        }
+
+        then {
+            assertAll (
+                { assert process.success },
+                { assert snapshot(
+                    process.out.rpt.collect{ meta,rpt -> file(rpt).name }, //assert unstable file
+                    process.out.index_html.collect{ meta,index_html -> file(index_html).name }, //assert unstable file
+                    process.out.heat_map_corr_plot,
+                    process.out.report_tsvs_ref,
+                    process.out.report_htmls_ref.collect{ meta,report_htmls_ref -> file(report_htmls_ref).name }, //assert unstable file
+                    process.out.report_tsvs_target,
+                    process.out.report_htmls_target.collect{ meta,report_htmls_target -> file(report_htmls_target).name }, //assert unstable file
+                    process.out.ranked_gene_list,
+                    process.out.gene_set_sizes,
+                    process.out.histogram.collect{ meta,histogram -> file(histogram).name }, //assert unstable file
+                    process.out.heatmap.collect{ meta,heatmap -> file(heatmap).name }, //assert unstable file
+                    process.out.pvalues_vs_nes_plot.collect{ meta,pvalues_vs_nes_plot -> file(pvalues_vs_nes_plot).name }, //assert unstable file
+                    process.out.ranked_list_corr.collect{ meta,ranked_list_corr -> file(ranked_list_corr).name }, //assert unstable file
+                    process.out.butterfly_plot.collect{ meta,butterfly_plot -> file(butterfly_plot).name }, //assert unstable file
+                    process.out.gene_set_tsv,
+                    process.out.gene_set_html[0][1].collect{ file(it).name }, //assert unstable file
+                    process.out.gene_set_heatmap[0][1].collect{ file(it).name }, // unstable file and unstable name
+                    process.out.snapshot[0][1].collect{ file(it).name }, //assert unstable file
+                    process.out.gene_set_enplot[0][1].collect{ file(it).name }, // unstable file and unstable name
+                    process.out.gene_set_dist[0][1].collect{ file(it).name }, //assert unstable file
+                    process.out.archive,
+                    process.out.versions
+                    ).match()
+                }
+            )
+        }
+    }
+
+    test("stub") {
+
+        options "-stub"
+
+        when {
+            process {
+                """
+                input[0] = [['id':'Condition_genotype_WT_KO', 'variable':'Condition genotype', 'reference':'WT', 'target':'KO', 'blocking':'batch'], file("https://github.com/nf-core/test-datasets/raw/refs/heads/modules/data/genomics/mus_musculus/gene_set_analysis/Condition_treatment_Control_Treated.gct", checkIfExists:true), file("https://github.com/nf-core/test-datasets/raw/refs/heads/modules/data/genomics/mus_musculus/gene_set_analysis/Condition_genotype_WT_KO.cls", checkIfExists:true), file("https://github.com/nf-core/test-datasets/raw/refs/heads/modules/data/genomics/mus_musculus/gene_set_analysis/mh.all.v2022.1.Mm.symbols.gmt", checkIfExists:true)]
+                input[1] = ['WT', 'KO']
+                input[2] = file("https://github.com/nf-core/test-datasets/raw/refs/heads/modules/data/genomics/mus_musculus/gene_set_analysis/Mus_musculus.anno.feature_metadata.chip", checkIfExists:true)
+                """
+            }
+        }
+
+        then {
+            assertAll (
+                { assert process.success },
+                { assert snapshot(process.out).match() },
+            )
+        }
+    }
+}