Adding SLURM execution options (#54)

* separated slurm options into separate config

* allowed retries with increasing resources

* added default resource requests for SRA downloads and preprocessing

* added resource defaults for assembly

* added defaults for annotation, phageFinder, and validation alignment

* added defaults for reads binning and reads-based taxonomy classification

* added resource defaults for SMA, contig-based taxonomic classification

* updated templates

* fix job naming

---------

Co-authored-by: Andre Philippe Watson <[email protected]>

Author: aw-watson

workflows/Nextflow/configs/slurm.config

@@ -0,0 +1,36 @@
+//sets executor to slurm
+executor.name = "slurm"
+//controls job names
+executor.jobName = {"$task.process"}
+executor.queueSize=5
+process {
+    scratch = true
+    //options to provide for all processes submitted as cluster jobs (e.g. "--account xx")
+    clusterOptions = ""
+    //resource requirements for processes
+    withLabel: "tiny" {
+        cpus = {2 * task.attempt}
+        memory = {4.GB * task.attempt}
+    }
+    withLabel: "small" {
+        cpus = {4* task.attempt}
+        memory = {8.GB * task.attempt}
+    }
+    withLabel: "medium" {
+        cpus = {8 * task.attempt}
+        memory = {16.GB * task.attempt}
+        maxRetries = 2
+    }
+    withLabel: "large" {
+        cpus = {8 * task.attempt}
+        memory = {32.GB * task.attempt}
+        maxRetries = 2
+    }
+    time = {2.hours * task.attempt}
+    //if not labeled, uses the below settings
+    cpus = {4 * task.attempt}
+    memory = {8.GB * task.attempt}
+
+    errorStrategy='retry'
+    maxRetries=3
+}

workflows/Nextflow/modules/contigsTaxonomyAssignment/contigsTaxonomyAssignment.nf

@@ -5,6 +5,7 @@
 //base process. Takes a FASTA file containing contigs and performs taxonomic analysis with MICCR (https://github.com/chienchi/miccr).
 process contigTaxonomy {
     label 'cta'
+    label 'medium'
     containerOptions "--compat --cleanenv \
                         --bind=${settings["miccrDB"]}:/venv/database/miccrDB"
     publishDir(
@@ -34,6 +35,7 @@ process contigTaxonomy {
 //adds multi-level taxonomic classification to results file. Takes in a .ctg.tsv file produced by MICCR.
 process addLineage {
     label 'cta'
+    label 'tiny'
     containerOptions "--compat --cleanenv \
                         --bind=${settings["miccrDB"]}:/venv/database/miccrDB"
     publishDir(
@@ -59,6 +61,7 @@ process addLineage {
 //and a coverage table (see https://github.com/chienchi/miccr/blob/master/utils/README.md), or from workflow runReadsToContig 
 process plotAndTable {
     label 'cta'
+    label 'tiny'
     publishDir(
         path: "${settings["contigsTaxonomyOutDir"]}",
 	mode: 'copy'

workflows/Nextflow/modules/phageFinder/phageFinder.nf

@@ -3,6 +3,7 @@
 
 process phageFinderPrep {
     label 'phageFinder'
+    label 'tiny'
 
     input:
     path gff
@@ -21,6 +22,7 @@ process phageFinderPrep {
 
 process phageFinder {
     label 'phageFinder'
+    label 'small'
     publishDir(
         path: "${settings["phageFinderOutDir"]}",
         mode: 'copy',
@@ -46,6 +48,7 @@ process phageFinder {
 
 process summary {
     label 'phageFinder'
+    label 'tiny'
     publishDir(
         path: "${settings["phageFinderOutDir"]}",
         mode: 'copy'

workflows/Nextflow/modules/processProvidedContigs/processProvidedContigs.nf

@@ -2,6 +2,7 @@
 
 process processProvidedContigs {
     label "processContigs"
+    label "tiny"
     publishDir(
         path: "${settings["processProvidedContigsOutDir"]}",
 	    mode: 'copy'

workflows/Nextflow/modules/readsBinning/readsBinning.nf

@@ -1,5 +1,6 @@
 process runBinning {
     label 'binning'
+    label 'small'
     publishDir(
         path: "${settings["readBinningOutDir"]}",
         mode: 'copy'

workflows/Nextflow/modules/readsTaxonomyAssignment/readsTaxonomyAssignment.nf

@@ -3,6 +3,7 @@
 //main RTA process
 process readsTaxonomy {
     label 'rta'
+    label 'large'
 
     containerOptions "--compat --cleanenv \
                         --bind=${settings["baseDB"]}:/venv/bin/../../../database \
@@ -45,6 +46,7 @@ process readsTaxonomy {
 //creates RTA config file based on input settings
 process readsTaxonomyConfig {
     label 'rta'
+    label 'small'
 
     containerOptions "--compat --cleanenv \
                         --bind=${settings["baseDB"]}:/venv/bin/../../../database \

workflows/Nextflow/modules/runAnnotation/runAnnotation.nf

@@ -3,6 +3,7 @@
 //sets taxonomic kingdom for analysis if none provided
 process autodetectKingdom {
     label 'annotation'
+    label 'tiny'
     containerOptions '--compat --bind .:/venv/bin/ec_info'
     input:
     path contigs
@@ -23,6 +24,7 @@ process autodetectKingdom {
 //process to invocate prokka
 process prokkaAnnotate {
     label 'annotation'
+    label 'small'
     containerOptions '--compat --bind .:/venv/bin/ec_info'
     publishDir(
         path: "${settings["annotationOutDir"]}",
@@ -86,6 +88,7 @@ process prokkaAnnotate {
 //process to invocate RATT
 process rattAnnotate {
     label 'annotation'
+    label 'small'
     containerOptions '--compat --bind .:/venv/bin/ec_info'
     publishDir(
         path: "${settings["annotationOutDir"]}",
@@ -124,6 +127,7 @@ process rattAnnotate {
 //plots feature count, protein size distribution, etc.
 process annPlot {
     label 'annotation'
+    label 'tiny'
     containerOptions '--compat --bind .:/venv/bin/ec_info'
     publishDir(
         path: "${settings["annotationOutDir"]}",
@@ -150,6 +154,7 @@ process annPlot {
 //generates KEGG pathway plots
 process keggPlots {
     label 'annotation'
+    label 'tiny'
     containerOptions '--compat --bind .:/venv/bin/ec_info'
     publishDir(
         path: "${settings["annotationOutDir"]}",

workflows/Nextflow/modules/runAntiSmash/runAntiSmash.nf

@@ -2,6 +2,7 @@
 //runs ANTISMASH on provided .fa or .gbk input.
 process antismash {
     label 'sma'
+    label 'medium'
     containerOptions "--bind ${settings["database"]}:/venv/database/antiSMASH"
     publishDir(
         path: "${settings["smaOutDir"]}",

workflows/Nextflow/modules/runAssembly/runAssembly.nf

@@ -7,6 +7,7 @@
 //main process for assembly with IDBA
 process idbaUD {
     label "assembly"
+    label "small"
     publishDir (
     path:"${settings["assemblyOutDir"]}",
     mode: 'copy',
@@ -81,6 +82,7 @@ process idbaUD {
 //prep for idba
 process idbaExtractLong {
     label "assembly"
+    label "tiny"
 
     input:
     path paired
@@ -105,6 +107,7 @@ process idbaExtractLong {
 //prep for idba
 process idbaPrepReads {
     label "assembly"
+    label "tiny"
     input:
     path paired
     path unpaired
@@ -128,6 +131,7 @@ process idbaPrepReads {
 //assemble using spades
 process spades {
     label "assembly"
+    label "small"
 
     publishDir (
     path: "${settings["assemblyOutDir"]}", 
@@ -212,8 +216,9 @@ process spades {
 
 //assemble using megahit
 process megahit {
-
     label "assembly"
+    label "small"
+
     publishDir(
     path: "${settings["assemblyOutDir"]}", 
     mode: 'copy',
@@ -272,6 +277,7 @@ process megahit {
 //assembly using unicycler
 process unicycler {
     label "assembly"
+    label "small"
     publishDir (
         path: "${settings["assemblyOutDir"]}", 
         mode: 'copy',
@@ -328,6 +334,7 @@ process unicycler {
 //filter long reads for unicycler
 process unicyclerPrep {
     label "assembly"
+    label "tiny"
 
     input:
     val settings
@@ -351,6 +358,7 @@ process unicyclerPrep {
 //assembly using lrasm
 process lrasm {
     label "assembly"
+    label "small"
 
     publishDir (
         path: "${settings["assemblyOutDir"]}", 
@@ -429,6 +437,8 @@ process lrasm {
 
 process renameFilterFasta {
     label "assembly"
+    label "small"
+
     publishDir(
         path: "${settings["assemblyOutDir"]}",
         mode: 'copy'
@@ -461,6 +471,7 @@ process renameFilterFasta {
 
 process bestIncompleteAssembly {
     label "assembly"
+    label "tiny"
     input:
 
     val x

workflows/Nextflow/modules/runReadsToContig/runReadsToContig.nf

@@ -2,6 +2,7 @@
 
 process validationAlignment {
     label 'r2c'
+    label 'small'
     publishDir(
         path: "${settings["assemblyOutDir"]}/readsMappingToContig",
         mode: 'copy'
@@ -85,6 +86,7 @@ process validationAlignment {
 
 process makeJSONcoverageTable {
     label 'r2c'
+    label 'tiny'
     publishDir(
         path: "${settings["assemblyOutDir"]}/readsMappingToContig",
         mode: 'copy',
@@ -118,6 +120,7 @@ process makeJSONcoverageTable {
 
 process extractUnmapped {
     label 'r2c'
+    label 'small'
     publishDir(
         path:"${settings["assemblyOutDir"]}/readsMappingToContig/",
         mode: 'copy',

workflows/Nextflow/modules/sra2fastq/sra2fastq.nf

@@ -6,6 +6,7 @@
 
 process sraDownload {
     label "sra2fastq"
+    label "small"
     tag "$accession"
     publishDir "${settings["outDir"]}/SRA_Download", mode: 'copy'
     containerOptions "--no-home"

workflows/Nextflow/nextflow.config

@@ -241,7 +241,3 @@ process {
 executor {
     submitRateLimit = '1/5sec'
 }
-
-//cleanup
-cleanup = false
-

workflows/Nextflow/templates/SRAdownload_config.tmpl

@@ -245,4 +245,5 @@ executor {
     submitRateLimit = '1/5sec'
 }
 
-includeConfig "<%= report_config %>"
+includeConfig "<%= report_config %>"
+includeConfig "<%= slurm_config %>"

workflows/Nextflow/templates/runFaQCs_config.tmpl

@@ -242,4 +242,5 @@ executor {
     submitRateLimit = '1/5sec'
 }
 
-includeConfig "<%= report_config %>"
+includeConfig "<%= report_config %>"
+includeConfig "<%= slurm_config %>"