Add min_identity and min_coverage params

README.md

@@ -24,10 +24,14 @@ kma index -i <your_scheme.fa> -o <your_scheme>
 ```
 nextflow run BCCDC-PHL/kma-cgmlst \
   --fastq_input </path/to/fastqs> \
+  [--min_identity <min_percent_identity>] \
+  [--min_coverage <min_percent_coverage>] \
   --scheme </path/to/cgmlst_scheme> \
   --outdir </path/to/output_dir> 
 ```
 
+The `--min_identity` and `--min_coverage` flags can be used to control the identity and coverage thresholds that are used to call an allele. They both default to 100% if the flags are omitted.
+
 Alternatively, a `samplesheet.csv` file can be provided, with fields: `ID`,`R1`,`R2`:
 
 ```
@@ -55,4 +59,4 @@ nextflow run BCCDC-PHL/kma-cgmlst \
   --scheme </path/to/cgmlst_scheme> \
   --versioned_outdir \
   --outdir </path/to/output_dir> 
-```
+```

bin/kma_result_to_mlst.py

@@ -107,15 +107,15 @@ def main(args):
                 str(alleles[best_allele]['depth']),
             ]))
 
-            if alleles[best_allele]['template_identity'] == 100 and alleles[best_allele]['template_coverage'] == 100:
+            if alleles[best_allele]['template_identity'] >= args.min_identity and alleles[best_allele]['template_coverage'] >= args.min_coverage:
                 mlst_output[alleles[best_allele]['locus_id']] = alleles[best_allele]['allele_id']
             else:
                 mlst_output[alleles[best_allele]['locus_id']] = '-'
         else:
                 mlst_output[locus] = '-'
 
 
-    with open(args.o, 'w') as f:
+    with open(args.output, 'w') as f:
         fieldnames = ['sample_id'] + all_loci
         writer = csv.DictWriter(f, fieldnames=fieldnames)
         writer.writeheader()
@@ -128,6 +128,8 @@ def main(args):
     parser.add_argument("--locus-allele-delimiter", help="Delimiter separating locus id from allele id", default='_')
     parser.add_argument("-s", "--sample-id", help="Sample ID", default='unknown')
     parser.add_argument("-a", "--alleles", help="List of all alleles")
-    parser.add_argument("-o")
+    parser.add_argument("-i", "--min-identity", type=float, default=100.0, help="Minimum identity to consider an allele match")
+    parser.add_argument("-c", "--min-coverage", type=float, default=100.0, help="Minimum coverage to consider an allele match")
+    parser.add_argument("-o", "--output", help="Output")
     args = parser.parse_args()
     main(args)

modules/kma_result_to_mlst.nf

@@ -20,6 +20,8 @@ process kma_result_to_mlst {
       --alleles ${scheme}.name \
       --sample-id "${sample_id}" \
       --locus-allele-delimiter "_" \
+      --min-identity ${params.min_identity} \
+      --min-coverage ${params.min_coverage} \
       -o ${sample_id}_cgmlst.csv \
       > ${sample_id}_locus_qc.csv
     """

nextflow.config

@@ -1,7 +1,7 @@
 manifest {
   author = 'Dan Fornika <[email protected]>, Sherrie Wang <[email protected]>'
   name = 'BCCDC-PHL/kma-cgmlst'
-  version = '0.1.0'
+  version = '0.1.1'
   description = 'BCCDC-PHL kma cgMLST'
   mainScript = 'main.nf'
   nextflowVersion = '>=20.01.0'
@@ -16,6 +16,8 @@ params {
   versioned_outdir = false
   pipeline_short_name = parsePipelineName(manifest.toMap().get('name'))
   pipeline_minor_version = parseMinorVersion(manifest.toMap().get('version'))
+  min_identity = 100.0
+  min_coverage = 100.0
 }
 
 def makeFastqSearchPath ( illumina_suffixes, fastq_exts ) {
@@ -48,7 +50,10 @@ profiles {
 }
 
 process {
-  withName: kma_align { 
+  withName: fastp { 
     cpus = 4
   }
+  withName: kma_align { 
+    cpus = 8
+  }
 }