Add Sage Bionetworks nf-core configuration

.github/workflows/main.yml

@@ -77,6 +77,7 @@ jobs:
           - "phoenix"
           - "prince"
           - "rosalind"
+          - "sage"
           - "sahmri"
           - "sanger"
           - "seg_globe"

README.md

@@ -131,6 +131,7 @@ Currently documentation is available for the following systems:
 - [PHOENIX](docs/phoenix.md)
 - [PRINCE](docs/prince.md)
 - [ROSALIND](docs/rosalind.md)
+- [SAGE BIONETWORKS](docs/sage.md)
 - [SANGER](docs/sanger.md)
 - [SEG_GLOBE](docs/seg_globe.md)
 - [UCT_HPC](docs/uct_hpc.md)

conf/sage.config

@@ -0,0 +1,58 @@
+params {
+  config_profile_description = 'The Sage Bionetworks profile'
+  config_profile_contact = 'Bruno Grande (@BrunoGrandePhD)'
+  config_profile_url = 'https://github.com/Sage-Bionetworks-Workflows'
+}
+
+process {
+
+  cpus   = { check_max( 1    * slow(task.attempt), 'cpus'   ) }
+  memory = { check_max( 6.GB * task.attempt,       'memory' ) }
+  time   = { check_max( 24.h * task.attempt,       'time'   ) }
+
+  errorStrategy = { task.exitStatus in [143,137,104,134,139,247] ? 'retry' : 'finish' }
+  maxRetries    = 5
+  maxErrors     = '-1'
+
+  // Process-specific resource requirements
+  withLabel:process_low {
+    cpus   = { check_max( 4     * slow(task.attempt),  'cpus'   ) }
+    memory = { check_max( 12.GB * task.attempt,        'memory' ) }
+    time   = { check_max( 24.h  * task.attempt,        'time'   ) }
+  }
+  withLabel:process_medium {
+    cpus   = { check_max( 12    * slow(task.attempt), 'cpus'   ) }
+    memory = { check_max( 36.GB * task.attempt,       'memory' ) }
+    time   = { check_max( 48.h  * task.attempt,       'time'   ) }
+  }
+  withLabel:process_high {
+    cpus   = { check_max( 24    * slow(task.attempt), 'cpus'   ) }
+    memory = { check_max( 72.GB * task.attempt,       'memory' ) }
+    time   = { check_max( 96.h  * task.attempt,       'time'   ) }
+  }
+  withLabel:process_long {
+    time   = { check_max( 192.h  * task.attempt,   'time'   ) }
+  }
+  withLabel:process_high_memory {
+    memory = { check_max( 200.GB * task.attempt,   'memory' ) }
+  }
+
+}
+
+aws {
+  region = "us-east-1"
+}
+
+params {
+  igenomes_base = 's3://sage-igenomes/igenomes'
+  max_memory    = 512.GB
+  max_cpus      = 64
+  max_time      = 168.h  // One week
+}
+
+// Function to slow the increase of the resource multipler
+// as attempts are made. The rationale is that some CPUs
+// don't need to be increased as fast as memory.
+def slow(attempt, factor = 2) {
+  return Math.ceil( attempt / factor) as int
+}

docs/sage.md

@@ -0,0 +1,24 @@
+# nf-core/configs: Sage Bionetworks Configuration
+
+To use this custom configuration, run the pipeline with `-profile sage`. This will download and launch the [`sage.config`](../conf/sage.config), which contains a number of optimizations relevant to Sage employees running workflows on AWS (_e.g._ using Nextflow Tower). These include:
+
+- Updating the default value for `igenomes_base` to `s3://sage-igenomes`
+- Increasing the default time limits because we run pipelines on AWS
+- Enabling retries by default when exit codes relate to insufficient memory
+- Allow pending jobs to finish if the number of retries are exhausted
+- Slowing the increase in the number of allocated CPU cores on retries
+
+## Additional information about iGenomes
+
+The following iGenomes prefixes have been copied from `s3://ngi-igenomes/` (`eu-west-1`) to `s3://sage-igenomes` (`us-east-1`). See [this script](https://github.com/Sage-Bionetworks-Workflows/nextflow-infra/blob/main/bin/mirror-igenomes.sh) for more information. The `sage-igenomes` S3 bucket has been configured to openly available, but files cannot be downloaded out of `us-east-1` to avoid egress charges. You can check the `conf/igenomes.config` file in each nf-core pipeline to figure out the mapping between genome IDs (_i.e._ for `--genome`) and iGenomes prefixes ([example](https://github.com/nf-core/rnaseq/blob/89bf536ce4faa98b4d50a8ec0a0343780bc62e0a/conf/igenomes.config#L14-L26)).
+
+- **Human Genome Builds**
+  - `Homo_sapiens/Ensembl/GRCh37`
+  - `Homo_sapiens/GATK/GRCh37`
+  - `Homo_sapiens/UCSC/hg19`
+  - `Homo_sapiens/GATK/GRCh38`
+  - `Homo_sapiens/NCBI/GRCh38`
+  - `Homo_sapiens/UCSC/hg38`
+- **Mouse Genome Builds**
+  - `Mus_musculus/Ensembl/GRCm38`
+  - `Mus_musculus/UCSC/mm10`

nfcore_custom.config

@@ -59,6 +59,7 @@ profiles {
   phoenix      { includeConfig "${params.custom_config_base}/conf/phoenix.config" }
   prince       { includeConfig "${params.custom_config_base}/conf/prince.config" }
   rosalind     { includeConfig "${params.custom_config_base}/conf/rosalind.config" }
+  sage         { includeConfig "${params.custom_config_base}/conf/sage.config" }
   sahmri       { includeConfig "${params.custom_config_base}/conf/sahmri.config" }
   sanger       { includeConfig "${params.custom_config_base}/conf/sanger.config"}
   seg_globe    { includeConfig "${params.custom_config_base}/conf/seg_globe.config"}