simplify differential_functional_enrichment outputs

Author: suzannejin

subworkflows/nf-core/differential_functional_enrichment/main.nf

@@ -33,7 +33,9 @@ workflow DIFFERENTIAL_FUNCTIONAL_ENRICHMENT {
 
     main:
 
-    ch_versions = channel.empty()
+    ch_versions  = channel.empty()
+    ch_results   = channel.empty()
+    ch_artifacts = channel.empty()
 
     // Add method information into meta map of ch_input
     // This information is used later to determine which method to run for each input
@@ -82,21 +84,40 @@ workflow DIFFERENTIAL_FUNCTIONAL_ENRICHMENT {
         .combine(ch_contrasts_transposed, by:0)
         .multiMap(criteria)
 
-
     // ----------------------------------------------------
     // Perform enrichment analysis with gprofiler2
     // ----------------------------------------------------
 
+    // run gprofiler2
     GPROFILER2_GOST(
         ch_input_for_other.input.filter{ index -> index[0].functional_method == 'gprofiler2' },
         ch_input_for_other.genesets.filter{ index -> index[0].functional_method == 'gprofiler2'},
         ch_input_for_other.background.filter{ index -> index[0].functional_method == 'gprofiler2'}
     )
 
+    // collect main results
+    ch_results = ch_results
+        .mix(
+            GPROFILER2_GOST.out.plot_html
+                .join(GPROFILER2_GOST.out.all_enrich, remainder: true)
+                .join(GPROFILER2_GOST.out.sub_enrich, remainder: true)
+        )
+    
+    // other results
+    ch_artifacts = ch_artifacts
+        .mix(
+            GPROFILER2_GOST.out.plot_png
+                .mix(GPROFILER2_GOST.out.sub_plot)
+                .mix(GPROFILER2_GOST.out.rds)
+                .mix(GPROFILER2_GOST.out.filtered_gmt)
+                .mix(GPROFILER2_GOST.out.session_info)
+        )
+
     // ----------------------------------------------------
     // Perform enrichment analysis with GSEA
     // ----------------------------------------------------
 
+    // prepare GSEA input files (gct, cls, chip)
     CUSTOM_TABULARTOGSEAGCT(ch_input_for_gsea.input)
     CUSTOM_TABULARTOGSEACLS(ch_input_for_gsea.contrasts_and_samples)
     CUSTOM_TABULARTOGSEACHIP(
@@ -117,32 +138,87 @@ workflow DIFFERENTIAL_FUNCTIONAL_ENRICHMENT {
             [meta, chip]
         }
 
+    // run gsea
     GSEA_GSEA(
         ch_in_gsea.input,
         ch_in_gsea.ref_target,
         ch_in_gsea.chip
     )
 
+    // collect main results
+    ch_results = ch_results
+        .mix(
+            GSEA_GSEA.out.report_tsvs_ref
+                .join(GSEA_GSEA.out.report_tsvs_target)
+        )
+
+    // other results
+    ch_artifacts = ch_artifacts
+        .mix(
+            GSEA_GSEA.out.rpt
+                .mix(GSEA_GSEA.out.index_html)
+                .mix(GSEA_GSEA.out.heat_map_corr_plot)
+                .mix(GSEA_GSEA.out.report_tsvs_ref)
+                .mix(GSEA_GSEA.out.report_htmls_ref)
+                .mix(GSEA_GSEA.out.report_tsvs_target)
+                .mix(GSEA_GSEA.out.report_htmls_target)
+                .mix(GSEA_GSEA.out.ranked_gene_list)
+                .mix(GSEA_GSEA.out.gene_set_sizes)
+                .mix(GSEA_GSEA.out.histogram)
+                .mix(GSEA_GSEA.out.heatmap)
+                .mix(GSEA_GSEA.out.pvalues_vs_nes_plot)
+                .mix(GSEA_GSEA.out.ranked_list_corr)
+                .mix(GSEA_GSEA.out.butterfly_plot)
+                .mix(GSEA_GSEA.out.gene_set_tsv)
+                .mix(GSEA_GSEA.out.gene_set_html)
+                .mix(GSEA_GSEA.out.gene_set_heatmap)
+                .mix(GSEA_GSEA.out.snapshot)
+                .mix(GSEA_GSEA.out.gene_set_enplot)
+                .mix(GSEA_GSEA.out.gene_set_dist)
+                .mix(GSEA_GSEA.out.archive)
+        )
+
     // ----------------------------------------------------
     // Perform enrichment analysis with DECOUPLER
     // ----------------------------------------------------
 
+    // run decoupler
     DECOUPLER_DECOUPLER(
         ch_input_for_other.input.filter{ index -> index[0].functional_method == 'decoupler' },
         ch_input_for_other.genesets.filter{ index -> index[0].functional_method == 'decoupler'},
         ch_input_for_other.features.filter{ index -> index[0].functional_method == 'decoupler'}
             .map{ meta, features_sheet, _features_id, _features_symbol -> [meta, features_sheet] }
     )
 
+    // collect main results
+    ch_results = ch_results
+        .mix(
+            DECOUPLER_DECOUPLER.out.dc_estimate
+                .join(DECOUPLER_DECOUPLER.out.dc_pvals)
+        )
+
+    // other results
+    ch_artifacts = ch_artifacts
+        .mix(DECOUPLER_DECOUPLER.out.png)
+
     // ----------------------------------------------------
     // Perform enrichment analysis with GREA
     // ----------------------------------------------------
 
+    // run gene ratio enrichment analysis
     PROPR_GREA(
         ch_input_for_other.input.filter{ index -> index[0].functional_method == 'grea' },
         ch_input_for_other.genesets.filter{ index -> index[0].functional_method == 'grea' }
     )
 
+    // collect main results
+    ch_results = ch_results
+        .mix(PROPR_GREA.out.results)
+    
+    // ----------------------------------------------------
+    // Outputs
+    // ----------------------------------------------------
+
     // collect versions info
     ch_versions = ch_versions
         .mix(GPROFILER2_GOST.out.versions)
@@ -156,48 +232,12 @@ workflow DIFFERENTIAL_FUNCTIONAL_ENRICHMENT {
     // here we emit the outputs that will be useful afterwards in the
     // nf-core/differentialabundance pipeline
 
-    // gprofiler2-specific outputs
-    gprofiler2_plot_html  = GPROFILER2_GOST.out.plot_html
-    gprofiler2_all_enrich = GPROFILER2_GOST.out.all_enrich
-    gprofiler2_sub_enrich = GPROFILER2_GOST.out.sub_enrich
-    gprofiler2_artifacts = GPROFILER2_GOST.out.plot_png
-        .mix(GPROFILER2_GOST.out.sub_plot)
-        .mix(GPROFILER2_GOST.out.rds)
-        .mix(GPROFILER2_GOST.out.filtered_gmt)
-        .mix(GPROFILER2_GOST.out.session_info)
-
-    // gsea-specific outputs
-    gsea_report = GSEA_GSEA.out.report_tsvs_ref.join(GSEA_GSEA.out.report_tsvs_target)
-    gsea_artifacts = GSEA_GSEA.out.rpt
-        .mix(GSEA_GSEA.out.index_html)
-        .mix(GSEA_GSEA.out.heat_map_corr_plot)
-        .mix(GSEA_GSEA.out.report_tsvs_ref)
-        .mix(GSEA_GSEA.out.report_htmls_ref)
-        .mix(GSEA_GSEA.out.report_tsvs_target)
-        .mix(GSEA_GSEA.out.report_htmls_target)
-        .mix(GSEA_GSEA.out.ranked_gene_list)
-        .mix(GSEA_GSEA.out.gene_set_sizes)
-        .mix(GSEA_GSEA.out.histogram)
-        .mix(GSEA_GSEA.out.heatmap)
-        .mix(GSEA_GSEA.out.pvalues_vs_nes_plot)
-        .mix(GSEA_GSEA.out.ranked_list_corr)
-        .mix(GSEA_GSEA.out.butterfly_plot)
-        .mix(GSEA_GSEA.out.gene_set_tsv)
-        .mix(GSEA_GSEA.out.gene_set_html)
-        .mix(GSEA_GSEA.out.gene_set_heatmap)
-        .mix(GSEA_GSEA.out.snapshot)
-        .mix(GSEA_GSEA.out.gene_set_enplot)
-        .mix(GSEA_GSEA.out.gene_set_dist)
-        .mix(GSEA_GSEA.out.archive)
-
-    // decoupler-specific outputs
-    decoupler_dc_estimate = DECOUPLER_DECOUPLER.out.dc_estimate
-    decoupler_dc_pvals = DECOUPLER_DECOUPLER.out.dc_pvals
-    decoupler_png = DECOUPLER_DECOUPLER.out.png
-
-    // grea-specific outputs
-    grea_results          = PROPR_GREA.out.results
+    // main results
+    results = ch_results
+
+    // artifacts - these are all the other files produced by the modules that are not the main results, but that we still want to keep and make available
+    artifacts = ch_artifacts
 
     // tool versions
-    versions              = ch_versions
+    versions = ch_versions
 }

subworkflows/nf-core/differential_functional_enrichment/meta.yml

@@ -92,136 +92,41 @@ input:
 
 output:
   - versions:
-      type: file
       description: |
-        File containing software versions
-        Structure: [ path(versions.yml) ]
-      pattern: "versions.yml"
-  - gprofiler2_all_enrich:
-      description: Channel containing the main enrichment table output from gprofiler2
+        File containing software versions.
       structure:
-        - meta:
-            type: map
-            description: Metadata map
-        - all_enrich:
-            type: file
-            description: |
-              Table listing all enriched pathways that were found by gprofiler2.
-              It can be empty, if none is found.
-            pattern: "*.gprofiler2.all_enriched_pathways.tsv"
-  - gprofiler2_sub_enrich:
-      description: Channel containing the secondary enrichment table output from gprofiler2.
-      structure:
-        - meta:
-            type: map
-            description: Metadata map
-        - sub_enrich:
-            type: file
-            description: |
-              Table listing enriched pathways that were found from one particular source.
-              Note that it will only be created if any were found.
-            pattern: "*.gprofiler2.*.sub_enriched_pathways.tsv"
-  - gprofiler2_plot_html:
-      description: Channel containing the html report generated from gprofiler2.
-      structure:
-        - meta:
-            type: map
-            description: Metadata map
-        - plot_html:
+        - versions.yml:
             type: file
-            description: |
-              HTML file; interactive Manhattan plot of all enriched pathways.
-              Note that this file will only be generated if enriched pathways were found.
-            pattern: "*.gprofiler2.gostplot.html"
-  - gprofiler2_artifacts:
-      description: Channel containing additional gprofiler2 artifacts.
-      structure:
-        - meta:
-            type: map
-            description: Metadata map
-        - artifact:
-            type: file
-            description: |
-              Additional outputs from gprofiler2, including gost results (RDS),
-              filtered GMT files, R session logs, and PNG plot files.
-            pattern: "*.{rds,gmt,log,png}"
-  - gsea_report:
-      description: Channel containing all the output from GSEA needed for further reporting.
+            description: File containing software versions
+            pattern: "versions.yml"
+  - results:
+      description: |
+        Channel containing the main results from all functional enrichment methods.
+        For gprofiler2, this includes the plot HTML, all enriched pathways table, and sub enriched pathways table.
+        For GSEA, this includes the report TSVs for reference and target groups.
+        For decoupler, this includes the estimate and p-values files.
+        For GREA, this includes the main results TSV.
       structure:
         - meta:
             type: map
             description: Metadata map
-        - reports_ref:
-            type: file
-            description: Main TSV results report file for the reference group.
-            pattern: "*gsea_report_for_${reference}.tsv"
-        - reports_target:
+        - results:
             type: file
-            description: Main TSV results report file for the target group.
-            pattern: "*gsea_report_for_${target}.tsv"
-  - gsea_artifacts:
-      description: Channel containing additional GSEA output artifacts.
+            description: Main result files from the functional enrichment method.
+  - artifacts:
+      description: |
+        Channel containing additional output files from all functional enrichment methods
+        that are not the main results but are still useful to keep.
+        For gprofiler2, this includes PNG plots, RDS files, filtered GMT, and session info.
+        For GSEA, this includes reports, plots, archives, and supporting HTML/TSV outputs.
+        For decoupler, this includes PNG plot files.
       structure:
         - meta:
             type: map
             description: Metadata map
         - artifact:
             type: file
-            description: |
-              Additional files generated by GSEA, including reports, plots, archives,
-              and supporting HTML/TSV outputs.
-            pattern: "*.{rpt,html,tsv,png,zip}"
-  - decoupler_dc_estimate:
-      description: Channel containing decoupler estimate results.
-      structure:
-        - meta:
-            type: map
-            description: |
-              Groovy Map containing sample information
-              e.g. [ id:'test', single_end ]
-        - estimate_files:
-            type: file
-            description: |
-              Files containing the estimation results of the enrichment(s)
-            pattern: "*estimate_decoupler.tsv"
-  - decoupler_dc_pvals:
-      description: Channel containing decoupler pvals results.
-      structure:
-        - meta:
-            type: map
-            description: |
-              Groovy Map containing sample information
-              e.g. [ id:'test', single_end ]
-        - pvals_files:
-            type: file
-            description: |
-              Files containing the p-values associated to the estimation
-              results of the enrichment(s)
-            pattern: "*pvals_decoupler.tsv"
-  - decoupler_png:
-      description: Channel containing decoupler png results.
-      structure:
-        - meta:
-            type: map
-            description: |
-              Groovy Map containing sample information
-              e.g. [ id:'test', single_end ]
-        - plot_files:
-            type: file
-            description: |
-              Files containing the plots associated to the estimation
-              results of the enrichment(s)
-            pattern: "*estimate_decoupler_plot.png"
-  - grea_results:
-      description: Channel containing the output from GREA.
-      structure:
-        - meta:
-            type: map
-            description: Metadata map
-        - results:
-            type: file
-            description: Main TSV results file.
-            pattern: "*.grea.tsv"
+            description: Additional output files from the functional enrichment method.
 authors:
   - "@suzannejin"
   - "@bjlang"