Refactor: extract calibration params to meta

modules/local/diann/assemble_empirical_library/main.nf

@@ -19,6 +19,7 @@ process ASSEMBLE_EMPIRICAL_LIBRARY {
     path "empirical_library.*", emit: empirical_library
     path "assemble_empirical_library.log", emit: log
     path "versions.yml", emit: versions
+    path "diann_calibrated_params.csv", emit: calibrated_params
 
     when:
     task.ext.when == null || task.ext.when
@@ -83,6 +84,14 @@ process ASSEMBLE_EMPIRICAL_LIBRARY {
 
     cp report.log.txt assemble_empirical_library.log
 
+    val_mass_acc_ms2=\$(grep "Averaged recommended settings" assemble_empirical_library.log | cut -d ' ' -f 11 | tr -cd "[0-9.]")
+    val_mass_acc_ms1=\$(grep "Averaged recommended settings" assemble_empirical_library.log | cut -d ' ' -f 15 | tr -cd "[0-9.]")
+    val_scan_window=\$(grep "Averaged recommended settings" assemble_empirical_library.log | cut -d ' ' -f 19 | tr -cd "[0-9.]")
+    if [ -z "\$val_mass_acc_ms2" ]; then val_mass_acc_ms2="0"; fi
+    if [ -z "\$val_mass_acc_ms1" ]; then val_mass_acc_ms1="0"; fi
+    if [ -z "\$val_scan_window" ]; then val_scan_window="0"; fi
+    echo "\${val_mass_acc_ms2},\${val_mass_acc_ms1},\${val_scan_window}" > diann_calibrated_params.csv
+
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":
         DIA-NN: \$(diann 2>&1 | grep "DIA-NN" | grep -oP "\\d+\\.\\d+(\\.\\w+)*(\\.[\\d]+)?")

modules/local/diann/assemble_empirical_library/meta.yml

@@ -35,5 +35,9 @@ output:
       type: file
       description: File containing software version
       pattern: "versions.yml"
+  - calibrated_params:
+      type: file
+      description: A file containing mass_acc_ms2, mass_acc_ms1, and scan_window extracted from the DIA-NN log.
+      pattern: "diann_calibrated_params.csv"
 authors:
   - "@daichengxin"

modules/local/diann/individual_analysis/main.nf

@@ -8,7 +8,7 @@ process INDIVIDUAL_ANALYSIS {
         'docker.io/biocontainers/diann:v1.8.1_cv1' }"
 
     input:
-    tuple val(meta), path(ms_file), path(fasta), path(diann_log), path(library)
+    tuple val(meta), path(ms_file), path(fasta), path(library)
     path(diann_config)
 
     output:
@@ -44,19 +44,39 @@ process INDIVIDUAL_ANALYSIS {
         }
     }
 
-    scan_window = params.scan_window
-
-    if (params.mass_acc_automatic | params.scan_window_automatic) {
-        mass_acc_ms2 = "\$(cat ${diann_log} | grep \"Averaged recommended settings\" | cut -d ' ' -f 11 | tr -cd \"[0-9]\")"
-        scan_window = "\$(cat ${diann_log} | grep \"Averaged recommended settings\" | cut -d ' ' -f 19 | tr -cd \"[0-9]\")"
-        mass_acc_ms1 = "\$(cat ${diann_log} | grep \"Averaged recommended settings\" | cut -d ' ' -f 15 | tr -cd \"[0-9]\")"
-    } else if (meta['precursormasstoleranceunit'].toLowerCase().endsWith('ppm') && meta['fragmentmasstoleranceunit'].toLowerCase().endsWith('ppm')) {
+    if (params.mass_acc_automatic || params.scan_window_automatic) {
+        if (meta.mass_acc_ms2 != "0" && meta.mass_acc_ms2 != null) {
+            mass_acc_ms2 = meta.mass_acc_ms2
+            mass_acc_ms1 = meta.mass_acc_ms1
+            scan_window  = meta.scan_window
+        } 
+        else if (meta['fragmentmasstolerance']) {
+            mass_acc_ms2 = meta['fragmentmasstolerance']
+            mass_acc_ms1 = meta['precursormasstolerance']
+            scan_window  = params.scan_window
+        } 
+        else {
+            mass_acc_ms2 = params.mass_acc_ms2
+            mass_acc_ms1 = params.mass_acc_ms1
+            scan_window  = params.scan_window
+        }
+    } else if (meta['precursormasstoleranceunit']?.toLowerCase()?.endsWith('ppm') && meta['fragmentmasstoleranceunit']?.toLowerCase()?.endsWith('ppm')) {
         mass_acc_ms1 = meta["precursormasstolerance"]
         mass_acc_ms2 = meta["fragmentmasstolerance"]
     } else {
-        mass_acc_ms2 = "\$(cat ${diann_log} | grep \"Averaged recommended settings\" | cut -d ' ' -f 11 | tr -cd \"[0-9]\")"
-        scan_window = "\$(cat ${diann_log} | grep \"Averaged recommended settings\" | cut -d ' ' -f 19 | tr -cd \"[0-9]\")"
-        mass_acc_ms1 = "\$(cat ${diann_log} | grep \"Averaged recommended settings\" | cut -d ' ' -f 15 | tr -cd \"[0-9]\")"
+        if (meta.mass_acc_ms2 != "0" && meta.mass_acc_ms2 != null) {
+            mass_acc_ms2 = meta.mass_acc_ms2
+            mass_acc_ms1 = meta.mass_acc_ms1
+            scan_window  = meta.scan_window
+        } else if (meta['fragmentmasstolerance']) {
+            mass_acc_ms2 = meta['fragmentmasstolerance']
+            mass_acc_ms1 = meta['precursormasstolerance']
+            scan_window  = params.scan_window
+        } else {
+            mass_acc_ms2 = params.mass_acc_ms2
+            mass_acc_ms1 = params.mass_acc_ms1
+            scan_window  = params.scan_window
+        }
     }
 
     diann_no_peptidoforms = params.diann_no_peptidoforms ? "--no-peptidoforms" : ""

modules/local/diann/individual_analysis/meta.yml

@@ -10,10 +10,6 @@ tools:
       homepage: https://github.com/vdemichev/DiaNN
       documentation: https://github.com/vdemichev/DiaNN
 input:
-  - diann_log:
-      type: file
-      description: DIA-NN log file
-      pattern: "assemble_empirical_library.log"
   - empirical_library:
       type: file
       description: An empirical spectral library from the .quant files.

nextflow.config

@@ -85,6 +85,10 @@ params {
     random_preanalysis_seed      = 42
     empirical_assembly_ms_n      = 200
 
+    // DIA-NN: INDIVIDUAL_ANALYSIS
+    mass_acc_ms2    = 15
+    mass_acc_ms1    = 15
+
     // DIA-NN: FINAL_QUANTIFICATION — summarization & output
     pg_level                = 2
     species_genes           = false

nextflow_schema.json

@@ -321,6 +321,20 @@
                     "help_text": " Ideally, should be approximately equal to the average number of data points per peak",
                     "default": 8
                 },
+                "mass_acc_ms2": {
+                    "type": "number",
+                    "description": "Set the MS2 mass accuracy (tolerance) to a specific value in ppm.",
+                    "fa_icon": "fas fa-bullseye",
+                    "help_text": "If specified, this overrides the automatic calibration. Corresponds to the --mass-acc parameter in DIA-NN.",
+                    "default": 15
+                },
+                "mass_acc_ms1": {
+                    "type": "number",
+                    "description": "Set the MS1 mass accuracy (tolerance) to a specific value in ppm.",
+                    "fa_icon": "fas fa-bullseye",
+                    "help_text": "If specified, this overrides the automatic calibration. Corresponds to the --mass-acc-ms1 parameter in DIA-NN.",
+                    "default": 15
+                },
                 "performance_mode": {
                     "type": "boolean",
                     "description": "Set Low RAM & High Speed Mode for DIANN, including min-corr, corr-diff, and time-corr-only three parameters",

workflows/dia.nf

@@ -59,12 +59,30 @@ workflow DIA {
     }
 
     if (params.skip_preliminary_analysis) {
-        assembly_log = channel.fromPath(params.empirical_assembly_log)
-        empirical_library = channel.fromPath(params.diann_speclib)
-        indiv_fin_analysis_in = ch_file_preparation_results.combine(ch_searchdb)
-            .combine(assembly_log)
-            .combine(empirical_library)
-        empirical_lib = empirical_library
+        def log_file = params.empirical_assembly_log ? file(params.empirical_assembly_log) : null
+        def parsed_m2 = "0"
+        def parsed_m1 = "0"
+        def parsed_w  = "0"        
+        if (log_file && log_file.exists()) {
+            def matcher = log_file.text =~ /Mass accuracy = ([0-9.]+)ppm, MS1 accuracy = ([0-9.]+)ppm, Scan window = ([0-9.]+)/
+            if (matcher) {
+                parsed_m2 = matcher[0][1]
+                parsed_m1 = matcher[0][2]
+                parsed_w  = matcher[0][3]
+            }
+        }        
+        indiv_fin_analysis_in = ch_file_preparation_results
+            .combine(ch_searchdb)
+            .combine(speclib)
+            .map { meta_map, ms_file, fasta, library ->
+                def new_meta = meta_map + [
+                    mass_acc_ms2 : parsed_m2,
+                    mass_acc_ms1 : parsed_m1,
+                    scan_window  : parsed_w
+                ]
+                return [ new_meta, ms_file, fasta, library ]
+            }
+        empirical_lib = speclib
     } else {
         //
         // MODULE: PRELIMINARY_ANALYSIS
@@ -102,9 +120,17 @@ workflow DIA {
             .mix(ASSEMBLE_EMPIRICAL_LIBRARY.out.versions)
         indiv_fin_analysis_in = ch_file_preparation_results
             .combine(ch_searchdb)
-            .combine(ASSEMBLE_EMPIRICAL_LIBRARY.out.log)
             .combine(ASSEMBLE_EMPIRICAL_LIBRARY.out.empirical_library)
-
+            .combine(ASSEMBLE_EMPIRICAL_LIBRARY.out.calibrated_params)
+            .map { meta_map, ms_file, fasta, library, param_file ->
+                def values = param_file.text.trim().split(',')
+                def new_meta = meta_map + [
+                    mass_acc_ms2 : values[0],
+                    mass_acc_ms1 : values[1],
+                    scan_window  : values[2]
+                ]
+                return [ new_meta, ms_file, fasta, library ]
+            }
         empirical_lib = ASSEMBLE_EMPIRICAL_LIBRARY.out.empirical_library
     }