ignoring workdirs and nproc is a Process command parameter

.gitignore

@@ -1,8 +1,10 @@
 *-env
 __pycache__
 .nextflow*
-work
+work/
 output*
 .DS_Store
 singularity/
 dev/
+.apptainer/
+workdirs/

bin/run_plate.py

@@ -49,8 +49,8 @@ def main(arguments):
                         launching pipleine""")
     parser.add_argument('-c', '--check-inputs', action='store_true', default=False,
                         help='verify inputs and exit')
-    parser.add_argument('-p', '--profile', choices=['hutch'],
-                        default='hutch_singularity', help="""[%(default)s]""")
+    parser.add_argument('-p', '--profile', choices=['hutch_batch', 'hutch_apptainer'],
+                        default='hutch_apptainer', help="""[%(default)s]""")
 
     args = parser.parse_args(arguments)
     plate = args.plate

main.nf

@@ -177,7 +177,7 @@ process cmsearch_orientations {
 
     """
     python3 -c "from Bio import SeqIO;import gzip;SeqIO.write(SeqIO.parse(gzip.open('${R1}', 'rt'), 'fastq'), 'R1.fa', 'fasta')"
-    cmsearch -E 10.0 --cpu 4 --hmmonly --noali --tblout scores.txt ${model} R1.fa
+    cmsearch -E 10.0 --cpu ${params.nproc} --hmmonly --noali --tblout scores.txt ${model} R1.fa
     split_reads.py --counts counts.csv --cmsearch scores.txt ${sampleid} ${R1} ${R2}
     """
 }
@@ -198,7 +198,7 @@ process vsearch_orientations {
 
     """
     python3 -c "from Bio import SeqIO;import gzip;SeqIO.write(SeqIO.parse(gzip.open('${R1}', 'rt'), 'fastq'), 'R1.fa', 'fasta')"
-    vsearch --usearch_global R1.fa --db ${library} --id 0.75 --query_cov 0.8 --strand both --threads 4 --top_hits_only --userfields query+qstrand --userout hits.tsv
+    vsearch --usearch_global R1.fa --db ${library} --id 0.75 --query_cov 0.8 --strand both --threads ${params.nproc} --top_hits_only --userfields query+qstrand --userout hits.tsv
     split_reads.py --counts counts.csv --vsearch hits.tsv ${sampleid} ${R1} ${R2}
     """
 }
@@ -291,16 +291,17 @@ process dada_dereplicate {
 
     """
     dada2_dada.R ${R1} ${R2} \
+        --counts counts.csv \
+        --data dada.rds \
         --errors ${model} \
-        --sampleid ${sampleid} \
         --orientation ${orientation} \
+        --overlaps overlaps.csv \
+        --nthreads ${params.nproc} \
         --params ${dada_params} \
-        --data dada.rds \
+        --sampleid ${sampleid} \
         --seqtab seqtab.csv \
         --seqtab-r1 seqtab_r1.csv \
-        --seqtab-r2 seqtab_r2.csv \
-        --counts counts.csv \
-        --overlaps overlaps.csv
+        --seqtab-r2 seqtab_r2.csv
     get_unmerged.R dada.rds \
         --forward-seqs unmerged_F.fasta \
         --reverse-seqs unmerged_R.fasta

nextflow.config

@@ -9,6 +9,8 @@ params {
     // Docker image is hosted on github
     container = "ghcr.io/nhoffman/dada2-nf:2.0.3"
     min_reads = 1
+    // for Processes that use nproc also see queueSize and maxForks
+    nproc = 4
     work_dir = "work"
 }
 
@@ -126,7 +128,8 @@ profiles {
         apptainer {
             enabled = true
             autoMounts = true
-            cacheDir = '/fh/scratch/delete90/fredricks_d/lab/nhoffman/apptainer/cache/'
+            // Nextflow specific, different from APPTAINER_CACHEDIR
+            cacheDir = '.apptainer/'
         }
         params {
             output = 'output'