bigbio · ypriverol · May 6, 2022 · Apr 22, 2022 · Apr 22, 2022 · Apr 23, 2022
diff --git a/assets/multiqc_config.yml b/assets/multiqc_config.yml
@@ -15,3 +15,7 @@ export_plots: true
 custom_logo: "./nf-core-quantms_logo_light.png"
 custom_logo_url: "https://github.com/bigbio/quantms"
 custom_logo_title: "quantms"
+
+sp:
+  quantms/exp_design:
+    fn: "*_design.tsv"
diff --git a/bin/msstats_tmt.R b/bin/msstats_tmt.R
@@ -210,6 +210,11 @@ if(typeof(reference_norm) == 'character'){
     reference_norm <- char_to_boolean[reference_norm]
 }
 
+if (length(args)<12) {
+    # outputPrefix
+    args[12] <- './msstatsiso'
+}
+
 csv_input <- args[1]
 contrast_str <- args[2]
 control_str <- args[3]
@@ -241,5 +246,5 @@ if (l == 1) {
     test.MSstatsTMT <- groupComparisonTMT(contrast.matrix=contrast_mat, data=processed.quant)
 
     #TODO allow manual input (e.g. proteins of interest)
-    write.table(test.MSstatsTMT$ComparisonResult, file=paste0("msstatsiso_results.csv"), quote=FALSE, sep='\t', row.names = FALSE)
+    write.table(test.MSstatsTMT$ComparisonResult, file=paste0(args[12],"_comparisons.csv"), quote=FALSE, sep='\t', row.names = FALSE)
 }
diff --git a/modules/local/msstats/main.nf b/modules/local/msstats/main.nf
@@ -1,4 +1,5 @@
 process MSSTATS {
+    tag "$msstats_csv_input.Name"
     label 'process_medium'
 
     conda (params.enable_conda ? "bioconda::bioconductor-msstats=4.2.0" : null)
@@ -9,7 +10,7 @@ process MSSTATS {
     }
 
     input:
-    path out_msstats
+    path msstats_csv_input
 
     output:
     // The generation of the PDFs from MSstats are very unstable, especially with auto-contrasts.
@@ -25,13 +26,14 @@ process MSSTATS {
 
     """
     msstats_plfq.R \\
-        ${out_msstats} \\
+        ${msstats_csv_input} \\
         ${params.contrasts} \\
         "${ref_con}" \\
         ${params.msstats_remove_one_feat_prot} \\
         ${params.msstatslfq_removeFewMeasurements} \\
         ${params.msstatslfq_feature_subset_protein} \\
         ${params.msstatslfq_quant_summary_method} \\
+        ${msstats_csv_input.baseName} \\
         $args \\
         > msstats.log \\
         || echo "Optional MSstats step failed. Please check logs and re-run or do a manual statistical analysis."

diff --git a/modules/local/msstatstmt/main.nf b/modules/local/msstatstmt/main.nf
@@ -1,4 +1,5 @@
 process MSSTATSTMT {
+    tag "$msstatstmt_csv_input.Name"
     label 'process_medium'
 
     conda (params.enable_conda ? "bioconda::bioconductor-msstatstmt=2.2.0" : null)
@@ -9,7 +10,7 @@ process MSSTATSTMT {
     }
 
     input:
-    path out_msstats_tmt
+    path msstatstmt_csv_input
 
     output:
     // The generation of the PDFs from MSstatsTMT are very unstable, especially with auto-contrasts.
@@ -25,7 +26,7 @@ process MSSTATSTMT {
 
     """
     msstats_tmt.R \\
-        ${out_msstats_tmt} \\
+        ${msstatstmt_csv_input} \\
         ${params.contrasts} \\
         "${ref_con}" \\
         ${params.msstats_remove_one_feat_prot} \\
@@ -36,6 +37,7 @@ process MSSTATSTMT {
         ${params.msstatsiso_global_norm} \\
         ${params.msstatsiso_remove_norm_channel} \\
         ${params.msstatsiso_reference_normalization} \\
+        ${msstatstmt_csv_input.baseName} \\
         $args \\
         > msstats_tmt.log \\
         || echo "Optional MSstatsTMT step failed. Please check logs and re-run or do a manual statistical analysis."

diff --git a/modules/local/openms/consensusid/main.nf b/modules/local/openms/consensusid/main.nf
@@ -1,4 +1,5 @@
 process CONSENSUSID {
+    tag "$meta.id"
     label 'process_medium'
     // TODO could be easily parallelized
     label 'process_single_thread'

diff --git a/modules/local/openms/extractpsmfeatures/main.nf b/modules/local/openms/extractpsmfeatures/main.nf
@@ -1,4 +1,5 @@
 process EXTRACTPSMFEATURES {
+    tag "$meta.id"
     label 'process_very_low'
     label 'process_single_thread'
     label 'openms'

diff --git a/modules/local/openms/falsediscoveryrate/main.nf b/modules/local/openms/falsediscoveryrate/main.nf
@@ -1,4 +1,5 @@
 process FALSEDISCOVERYRATE {
+    tag "$meta.id"
     label 'process_low'
     label 'process_single_thread'
     label 'openms'

diff --git a/modules/local/openms/idfilter/main.nf b/modules/local/openms/idfilter/main.nf
@@ -1,5 +1,5 @@
 process IDFILTER {
-
+    tag {task.ext.suffix == ".idXML" ? "$meta.id" : "$id_file.baseName"}
     label 'process_very_low'
     label 'process_single_thread'
     label 'openms'

diff --git a/modules/local/openms/idpep/main.nf b/modules/local/openms/idpep/main.nf
@@ -1,4 +1,5 @@
 process IDPEP {
+    tag "$meta.id"
     label 'process_very_low'
 
     conda (params.enable_conda ? "bioconda::openms=2.8.0" : null)

diff --git a/modules/local/openms/indexpeptides/main.nf b/modules/local/openms/indexpeptides/main.nf
@@ -1,4 +1,5 @@
 process INDEXPEPTIDES {
+    tag "$meta.id"
     label 'process_low'
 
     conda (params.enable_conda ? "bioconda::openms=2.8.0" : null)

diff --git a/modules/local/openms/msstatsconverter/main.nf b/modules/local/openms/msstatsconverter/main.nf
@@ -1,4 +1,5 @@
 process MSSTATSCONVERTER {
+    tag "$exp_file.Name"
     label 'process_low'
 
     conda (params.enable_conda ? "bioconda::openms=2.8.0" : null)
@@ -24,7 +25,7 @@ process MSSTATSCONVERTER {
         -in ${consensusXML} \\
         -in_design ${exp_file} \\
         -method ${quant_method} \\
-        -out out_msstats.csv \\
+        -out ${exp_file.baseName}_out_msstats.csv \\
         $args \\
         |& tee MSstatsConverter.log
 

diff --git a/modules/local/openms/proteomicslfq/main.nf b/modules/local/openms/proteomicslfq/main.nf
@@ -1,4 +1,5 @@
 process PROTEOMICSLFQ {
+    tag "${expdes.baseName - ~/_design$/}"
     label 'process_high'
 
     conda (params.enable_conda ? "bioconda::openms=2.8.0" : null)
@@ -13,10 +14,10 @@ process PROTEOMICSLFQ {
     path(fasta)
 
     output:
-    path "out.mzTab", emit: out_mztab
-    path "out.consensusXML", emit: out_consensusXML
-    path "out_msstats.csv", emit: out_msstats optional true
-    path "out_triqler.tsv", emit: out_triqler optional true
+    path "${expdes.baseName - ~/_design$/}.mzTab", emit: out_mztab
+    path "${expdes.baseName - ~/_design$/}.consensusXML", emit: out_consensusXML
+    path "*out_msstats.csv", emit: out_msstats optional true
+    path "*out_triqler.tsv", emit: out_triqler optional true
     path "debug_mergedIDs.idXML", emit: debug_mergedIDs optional true
     path "debug_mergedIDs_inference.idXML", emit: debug_mergedIDs_inference optional true
     path "debug_mergedIDsGreedyResolved.idXML", emit: debug_mergedIDsGreedyResolved optional true
@@ -28,8 +29,8 @@ process PROTEOMICSLFQ {
 
     script:
     def args = task.ext.args ?: ''
-    def msstats_present = params.quantification_method == "feature_intensity" ? '-out_msstats out_msstats.csv' : ''
-    def triqler_present = (params.quantification_method == "feature_intensity") && (params.add_triqler_output) ? '-out_triqler out_triqler.tsv' : ''
+    def msstats_present = params.quantification_method == "feature_intensity" ? "-out_msstats ${expdes.baseName - ~/_design$/}_msstats_in.csv" : ""
+    def triqler_present = (params.quantification_method == "feature_intensity") && (params.add_triqler_output) ? "-out_triqler ${expdes.baseName - ~/_design$/}_triqler_in.tsv" : ""
     def decoys_present = (params.quantify_decoys || ((params.quantification_method == "feature_intensity") && params.add_triqler_output)) ? '-PeptideQuantification:quantify_decoys' : ''
 
     """
@@ -46,12 +47,12 @@ process PROTEOMICSLFQ {
         -protein_quantification ${params.protein_quant} \\
         -alignment_order ${params.alignment_order} \\
         -picked_proteinFDR true \\
-        -out out.mzTab \\
+        -out ${expdes.baseName - ~/_design$/}.mzTab \\
         -threads ${task.cpus} \\
         ${msstats_present} \\
         ${triqler_present} \\
         ${decoys_present} \\
-        -out_cxml out.consensusXML \\
+        -out_cxml ${expdes.baseName - ~/_design$/}.consensusXML \\
         -proteinFDR ${params.protein_level_fdr_cutoff} \\
         $args \\
         |& tee proteomicslfq.log

diff --git a/modules/local/openms/thirdparty/percolator/main.nf b/modules/local/openms/thirdparty/percolator/main.nf
@@ -1,4 +1,5 @@
 process PERCOLATOR {
+    tag "$meta.id"
     label 'process_medium'
 
     conda (params.enable_conda ? "bioconda::openms-thirdparty=2.8.0" : null)

diff --git a/modules/local/pmultiqc/main.nf b/modules/local/pmultiqc/main.nf
@@ -14,14 +14,14 @@ process PMULTIQC {
 
     output:
     path "*.html", emit: ch_pmultiqc_report
-    path "*.db", optional:true, emit: ch_pmultiqc_db
+    path "*.db", optional: true, emit: ch_pmultiqc_db
     path "versions.yml", emit: versions
     path "*_data", emit: data
-    path "*_plots", optional:true, emit: plots
+    path "*_plots", optional: true, emit: plots
 
     script:
     def args = task.ext.args ?: ''
-    def disable_pmultqic = params.enable_pmultiqc ? "": "--disable_plugin"
+    def disable_pmultqic = params.enable_pmultiqc ? "" : "--disable_plugin"
 
     """
     multiqc \\

diff --git a/modules/local/preprocess_expdesign.nf b/modules/local/preprocess_expdesign.nf
@@ -5,24 +5,25 @@
 process PREPROCESS_EXPDESIGN {
     label 'process_very_low'
     label 'process_single_thread'
+    tag "$design.Name"
 
     container "frolvlad/alpine-bash"
 
     input:
     path design
 
     output:
-    path "experimental_design.tsv", emit: ch_expdesign
-    path "config.tsv", emit: ch_config
+    path "${design.baseName}_design.tsv", emit: ch_expdesign
+    path "${design.baseName}_config.tsv", emit: ch_config
 
     script:
 
     """
     # since we know that we will need to convert from raw to mzML for all tools that need the design (i.e., OpenMS tools)
     # we edit the design here and change the endings.
-    sed 's/.raw\\t/.mzML\\t/I' $design > experimental_design.tsv
+    sed 's/.raw\\t/.mzML\\t/I' ${design} > ${design.baseName}_design.tsv
 
     # here we extract the filenames and fake an empty config (since the config values will be deduced from the workflow params)
-    a=\$(grep -n '^\$' $design | head -n1| awk -F":" '{print \$1}'); sed -e ''"\${a}"',\$d' $design > config.tsv
+    a=\$(grep -n '^\$' ${design} | head -n1| awk -F":" '{print \$1}'); sed -e ''"\${a}"',\$d' ${design} > ${design.baseName}_config.tsv
     """
 }
diff --git a/modules/local/sdrfparsing/main.nf b/modules/local/sdrfparsing/main.nf
@@ -1,4 +1,5 @@
 process SDRFPARSING {
+    tag "$sdrf.Name"
     label 'process_low'
 
     conda (params.enable_conda ? "conda-forge::pandas_schema bioconda::sdrf-pipelines=0.0.21" : null)
@@ -10,9 +11,9 @@ process SDRFPARSING {
     path sdrf
 
     output:
-    path "experimental_design.tsv", optional:true, emit: ch_expdesign
-    path "openms.tsv", optional:true, emit: ch_sdrf_config_file
-    path "*.xml", optional:true, emit: mqpar
+    path "${sdrf.baseName}_design.tsv", optional: true, emit: ch_expdesign
+    path "${sdrf.baseName}_config.tsv", optional: true, emit: ch_sdrf_config_file
+    path "*.xml", optional: true, emit: mqpar
     path "*.log", emit: log
     path "versions.yml", emit: version
 
@@ -24,7 +25,9 @@ process SDRFPARSING {
     ## -l for legacy behavior to always add sample columns
     ## TODO Update the sdrf-pipelines to dynamic print versions
 
-    parse_sdrf convert-openms -t2 -l -s ${sdrf} |& tee sdrf_parsing.log
+    parse_sdrf convert-openms -t2 -l -s ${sdrf} |& tee ${sdrf.baseName}_parsing.log
+    mv openms.tsv ${sdrf.baseName}_config.tsv
+    mv experimental_design.tsv ${sdrf.baseName}_design.tsv
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":