Add taxon_merge_strategy parameter to nextflow.config

Author: Joon-Klaps

conf/modules.config

@@ -478,6 +478,7 @@ process {
             }
 
             withName: KAIJU_MERGEOUTPUTS {
+                ext.args = {"-c ${params.taxon_merge_strategy}"}
                 publishDir =
                     [
                         path: { "${params.outdir}/assembly/polishing/intermediate/precluster/merged_classifications" },

nextflow.config

@@ -87,6 +87,7 @@ params {
     assembler_patterns          = null
     skip_precluster             = false
     keep_unclassified           = true
+    taxon_merge_strategy        = 'lowest'
 
     // > HYBRID CONSENSUS
     skip_hybrid_consensus       = false

nextflow_schema.json

@@ -552,20 +552,28 @@
                     "fa_icon": "fas fa-forward",
                     "description": "Skip the preclustering of assemblies to facilitate downstream processing of assemblies"
                 },
-                "cluster_method": {
-                    "type": "string",
-                    "default": "mash",
-                    "description": "Cluster algorithm used for contigs",
-                    "enum": ["cdhitest", "vsearch", "mmseqs-linclust", "mmseqs-cluster", "vrhyme", "mash", "diamond"],
-                    "fa_icon": "fas fa-bezier-curve"
-                },
                 "keep_unclassified": {
                     "type": "boolean",
                     "default": true,
                     "fa_icon": "fas fa-filter",
                     "description": "Keep the contigs that could not be classified with the taxonomic databases (`kaiju_db` & `kraken2_db`)",
                     "help_text": "Within the preclustering step, all contigs will get a taxonomic classification using the provided databases for the metagenomic tools. In some cases, the number of unclassified contigs, can be very large if the database is restrictive. This will result in large clusters in downstream processing that can take up a lot of resources despite not being a priority in some analyses. So set it to `True` if you want to keep unclassified contigs and set it to `False` if you don't want to keep them. "
                 },
+                "taxon_merge_strategy": {
+                    "type": "string",
+                    "default": "lowest",
+                    "fa_icon": "far fa-object-ungroup",
+                    "description": "Taxon conflict resolution mode, must be 1 (Kaiju), 2 (Kraken),  lca, or lowest.",
+                    "help_text": "The option -c determines the method of resolving conflicts in the taxonomic assignment for a read.\nPossible values are '1', '2', 'lca', 'lowest':\n  '1' -> the taxon id from Kaiju is used.\n  '2' -> the taxon id from Kraken is used.\n  'lca' -> the least common ancestor of the two taxon ids from both input files is used.\n  'lowest' -> the lower rank of the two taxa is used if they are within the same lineage. Otherwise the LCA is used.",
+                    "enum": ["1", "2", "lca", "lowest"]
+                },
+                "cluster_method": {
+                    "type": "string",
+                    "default": "mash",
+                    "description": "Cluster algorithm used for contigs",
+                    "enum": ["cdhitest", "vsearch", "mmseqs-linclust", "mmseqs-cluster", "vrhyme", "mash", "diamond"],
+                    "fa_icon": "fas fa-bezier-curve"
+                },
                 "network_clustering": {
                     "type": "string",
                     "default": "connected_components",