Explicitly declare plugins in nextflow.config

docs/how-to/offline.rst

@@ -7,9 +7,10 @@ pgsc_calc has been deployed on secure platforms like Trusted Research
 Environments (TREs). Running pgsc_calc is a little bit more difficult in this
 case. The basic set up approach is to:
 
-1. Download containers 
-2. Download reference data
-3. Download scoring files 
+1. Set up Nextflow
+2. Download containers 
+3. Download reference data
+4. Download scoring files 
 
 And transfer everything to your offline environment.  
 
@@ -22,6 +23,41 @@ if you are having problems and we'll try our best to help you.
 
 .. _open a discussion on Github: https://github.com/PGScatalog/pgsc_calc/discussions
 
+Set up Nextflow
+----------------
+
+From the Nextflow documentation for `offline usage <https://www.nextflow.io/docs/latest/plugins.html#offline-usage>`_:
+
+1. Run the test profile of the calculator with ``nextflow run pgscatalog/pgsc_calc -r test,<docker/singularity/conda>``
+
+.. tip::
+
+  It doesn't matter if the profile you use on your computer with internet access is different to the profile you use in the airlocked environment. 
+
+  The important thing is that Nextflow automatically configures itself using an internet connection. 
+
+2. Transfer the Nextflow binary and ``$NXF_HOME/.nextflow`` directory to your airlocked environment 
+
+.. tip::
+
+  ``$NXF_HOME`` is ``$HOME`` by default, so the directory is probably ``~/.nextflow``
+
+.. warning::
+
+  Make sure to transfer the Nextflow binary even if the airlocked environment already has Nextflow installed. It's important that the Nextflow versions match across both environments.
+
+3. Remember to always set the environment variable ``NXF_OFFLINE='true'`` in the offline environment
+
+
+.. tip::
+
+    You shouldn't need to:
+
+    1. Edit any Nextflow configuration files
+    2. Manually download any plugins 
+
+    Unless you want to use a special plugin in the airlocked environment 
+
 Preload container images
 ------------------------
 

lib/WorkflowMain.groovy

@@ -12,11 +12,12 @@ class WorkflowMain {
     public static String citation(workflow) {
         return "If you use ${workflow.manifest.name} for your analysis please cite:\n\n" +
             "* The Polygenic Score Catalog\n" +
+            "  https://doi.org/10.1101/2024.05.29.24307783\n" + 
             "  https://doi.org/10.1038/s41588-021-00783-5\n\n" +
             "* The nf-core framework\n" +
             "  https://doi.org/10.1038/s41587-020-0439-x\n\n" +
             "* Software dependencies\n" +
-            "  https://github.com/${workflow.manifest.name}/blob/master/CITATIONS.md"
+            "  https://github.com/${workflow.manifest.name}/blob/main/CITATIONS.md"
     }
 
 

main.nf

@@ -15,13 +15,13 @@ nextflow.enable.dsl = 2
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 */
 
+include { paramsHelp } from 'plugin/nf-schema'
+
 // Print help message if needed
 if (params.help) {
-    def logo = NfcoreTemplate.logo(workflow, params.monochrome_logs)
-    def citation = '\n' + WorkflowMain.citation(workflow) + '\n'
-    def String command = '\n' + "\$ nextflow run ${workflow.manifest.name} -profile test,docker" + '\n'
-    log.info logo + command + citation + NfcoreTemplate.dashedLine(params.monochrome_logs)
-    System.exit(0)
+    log.info paramsHelp("nextflow run pgscatalog/pgsc_calc --input input_file.csv")
+    log.info "See https://pgsc-calc.readthedocs.io/en/latest/getting-started.html for more help"
+    exit 0
 }
 
 WorkflowMain.initialise(workflow, params, log, args)

modules/local/match_variants.nf

@@ -23,7 +23,6 @@ process MATCH_VARIANTS {
 
     script:
     def args = task.ext.args                ?: ''
-    def fast = params.fast_match            ? '--fast'              : ''
     def ambig = params.keep_ambiguous       ? '--keep_ambiguous'    : ''
     def multi = params.keep_multiallelic    ? '--keep_multiallelic' : ''
     def match_chrom = meta.chrom.contains("ALL") ? '' : "--chrom $meta.chrom"
@@ -42,7 +41,6 @@ process MATCH_VARIANTS {
         $match_chrom \
         $ambig \
         $multi \
-        $fast \
         --outdir \$PWD \
         -v
 

nextflow.config

@@ -44,7 +44,6 @@ params {
     normalization_method = "empirical mean mean+var"
     n_normalization = 4
 
-
     // compatibility params
     liftover = false
     target_build = null
@@ -54,11 +53,9 @@ params {
     min_overlap = 0.75
     keep_ambiguous = false
     keep_multiallelic = false
-    fast_match = false
     copy_genomes = false
     genotypes_cache = null
 
-
     // Debug params
     only_bootstrap = false
     only_input = false
@@ -68,9 +65,6 @@ params {
     only_score = false
     skip_ancestry = true
 
-    // deprecated params
-    platform = null
-
     // Boilerplate options
     outdir                     = "$launchDir/results"
     publish_dir_mode           = 'copy'
@@ -96,14 +90,6 @@ params {
     max_memory                 = '128.GB'
     max_cpus                   = 16
     max_time                   = '240.h'
-
-    // Schema validation default options
-    validationFailUnrecognisedParams = false
-    validationLenientMode            = false
-    validationSchemaIgnoreParams     = 'genomes,igenomes_base,platform,only_bootstrap,only_input,only_compatible,only_match,only_score'
-    validationShowHiddenParams       = false
-    validate_params                  = true
-
 }
 
 // Load base.config by default for all pipelines
@@ -308,3 +294,27 @@ def check_max(obj, type) {
         }
     }
 }
+
+plugins { 
+    id 'nf-schema@2.0.0' // validation of parameters
+    id 'nf-prov@1.2.2'
+} 
+
+ prov {
+    enabled = true
+    formats {
+        bco {
+            file = "${params.outdir}/pipeline_info/manifest_${trace_timestamp}.bco.json"
+        }
+    }
+}
+
+validation {
+    // Schema validation default options
+    monochromeLogs         = params.monochrome_logs 
+    failUnrecognisedParams = false
+    lenientMode            = false
+    defaultIgnoreParams    = ['platform']
+    ignoreParams           = ['genomes','igenomes_base',',only_bootstrap','only_input','only_compatible','only_match','only_score']
+    showHiddenParams       = false
+}

nextflow_schema.json

@@ -1,10 +1,10 @@
 {
-  "$schema": "http://json-schema.org/draft-07/schema",
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
   "$id": "https://raw.githubusercontent.com/pgscatalog/pgsc_calc/master/nextflow_schema.json",
   "title": "pgscatalog/pgsc_calc pipeline parameters",
   "description": "This pipeline applies scoring files from the PGS Catalog to target set(s) of genotyped samples",
   "type": "object",
-  "definitions": {
+  "defs": {
     "input_output_options": {
       "title": "Input/output options",
       "type": "object",
@@ -31,11 +31,13 @@
         },
         "pgs_id": {
           "type": "string",
-          "description": "A comma separated list of PGS score IDs, e.g. PGS000802"
+          "description": "A comma separated list of PGS score IDs, e.g. PGS000802",
+          "pattern": "PGS[0-9]{6}"
         },
         "pgp_id": {
           "type": "string",
-          "description": "A comma separated list of PGS Catalog publications, e.g. PGP000001"
+          "description": "A comma separated list of PGS Catalog publications, e.g. PGP000001",
+          "pattern": "PGP[0-9]{6}"
         },
         "trait_efo": {
           "type": "string",
@@ -58,7 +60,8 @@
           "type": "string",
           "description": "Path to the output directory where the results will be saved.",
           "fa_icon": "fas fa-folder-open",
-          "format": "directory-path"
+          "format": "directory-path",
+          "default": "results"
         },
         "email": {
           "type": "string",
@@ -255,10 +258,6 @@
           "type": "boolean",
           "description": "Keep matches of scoring file variants to strand ambiguous variants (e.g. A/T and C/G SNPs) in the target dataset. This assumes the scoring file and target dataset report variants on the same strand."
         },
-        "fast_match": {
-          "type": "boolean",
-          "description": "Enable fast matching, which significantly increases RAM usage (32GB minimum recommended)"
-        },
         "min_overlap": {
           "type": "number",
           "default": 0.75,
@@ -353,38 +352,31 @@
           "description": "Do not use coloured log outputs.",
           "fa_icon": "fas fa-palette",
           "hidden": true
-        },
-        "validate_params": {
-          "type": "boolean",
-          "description": "Boolean whether to validate parameters against the schema at runtime",
-          "default": true,
-          "fa_icon": "fas fa-check-square",
-          "hidden": true
         }
       }
     }
   },
   "allOf": [
     {
-      "$ref": "#/definitions/input_output_options"
+      "$ref": "#/defs/input_output_options"
     },
     {
-      "$ref": "#/definitions/ancestry_options"
+      "$ref": "#/defs/ancestry_options"
     },
     {
-      "$ref": "#/definitions/reference_options"
+      "$ref": "#/defs/reference_options"
     },
     {
-      "$ref": "#/definitions/compatibility_options"
+      "$ref": "#/defs/compatibility_options"
     },
     {
-      "$ref": "#/definitions/matching_options"
+      "$ref": "#/defs/matching_options"
     },
     {
-      "$ref": "#/definitions/max_job_request_options"
+      "$ref": "#/defs/max_job_request_options"
     },
     {
-      "$ref": "#/definitions/generic_options"
+      "$ref": "#/defs/generic_options"
     }
   ],
   "properties": {
@@ -442,15 +434,6 @@
     },
     "config_profile_url": {
       "type": "string"
-    },
-    "validationFailUnrecognisedParams": {
-      "type": "boolean"
-    },
-    "validationLenientMode": {
-      "type": "boolean"
-    },
-    "validationShowHiddenParams": {
-      "type": "boolean"
     }
   }
 }

workflows/pgsc_calc.nf

@@ -4,7 +4,8 @@
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 */
 
-include { paramsSummaryLog; paramsSummaryMap } from 'plugin/nf-validation'
+include { validateParameters; paramsSummaryLog; paramsSummaryMap } from 'plugin/nf-schema'
+
 
 def logo = NfcoreTemplate.logo(workflow, params.monochrome_logs)
 def citation = '\n' + WorkflowMain.citation(workflow) + '\n'
@@ -15,6 +16,9 @@ log.info logo + paramsSummaryLog(workflow) + citation
 
 WorkflowPgscCalc.initialise(params, log)
 
+// new approach to validating parameters
+validateParameters()
+
 /*
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
     DEBUG OPTIONS TO HALT WORKFLOW EXECUTION
@@ -123,19 +127,6 @@ include { APPLY_SCORE          } from '../subworkflows/local/apply_score'
 include { REPORT               } from '../subworkflows/local/report'
 include { DUMPSOFTWAREVERSIONS } from '../modules/local/dumpsoftwareversions'
 
-
-/*
-========================================================================================
-    DEPRECATION WARNINGS
-========================================================================================
-*/
-
-if (params.platform) {
-    System.err.println "--platform has been deprecated to match nf-core framework"
-    System.err.println "Please use -profile docker,arm instead"
-    System.exit(1)
-}
-
 /*
 ========================================================================================
     RUN MAIN WORKFLOW