From 4ceb369af53ec89bce5dc4852ee28a366c2501d7 Mon Sep 17 00:00:00 2001 From: ameynert Date: Wed, 21 Jul 2021 12:10:43 +0100 Subject: [PATCH 1/5] eddie.config for viralrecon pipeline --- conf/pipeline/viralrecon/eddie.config | 9 +++++++++ 1 file changed, 9 insertions(+) create mode 100644 conf/pipeline/viralrecon/eddie.config diff --git a/conf/pipeline/viralrecon/eddie.config b/conf/pipeline/viralrecon/eddie.config new file mode 100644 index 0000000..38192ef --- /dev/null +++ b/conf/pipeline/viralrecon/eddie.config @@ -0,0 +1,9 @@ +process { + withName : '.*PICARD.*' { + clusterOptions = {"-l h_vmem=${(task.memory + 4.GB).bytes/task.cpus}"} + } + + withName : '.*SNPEFF.*' { + clusterOptions = {"-l h_vmem=${(task.memory + 4.GB).bytes/task.cpus}"} + } +} From 5224eb55742c7fc52b17d6d2938b97623d2a150b Mon Sep 17 00:00:00 2001 From: ameynert Date: Wed, 21 Jul 2021 12:15:17 +0100 Subject: [PATCH 2/5] Updated resources folder & institute name Local resources folder was re-organized & re-named BioinformaticsResources. Added note about viralrecon & atacseq pipeline-specific config. IGMM name-changed recently; updated to IGC. --- docs/eddie.md | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/docs/eddie.md b/docs/eddie.md index e5dbe5c..485e20a 100644 --- a/docs/eddie.md +++ b/docs/eddie.md @@ -1,10 +1,10 @@ # nf-core/configs: Eddie Configuration -nf-core pipelines sarek, rnaseq, and atacseq have all been tested on the University of Edinburgh Eddie HPC. +nf-core pipelines sarek, rnaseq, atacseq, and viralrecon have all been tested on the University of Edinburgh Eddie HPC. All except atacseq have pipeline-specific config files; atacseq does not yet support this. ## Getting help -There is a Slack channel dedicated to eddie users on the MRC IGMM Slack: [https://igmm.slack.com/channels/eddie3](https://igmm.slack.com/channels/eddie3) +There is a Slack channel dedicated to eddie users on the MRC IGC Slack: [https://igmm.slack.com/channels/eddie3](https://igmm.slack.com/channels/eddie3) ## Using the Eddie config profile @@ -35,13 +35,13 @@ This config enables Nextflow to manage the pipeline jobs via the SGE job schedul ## Singularity set-up -Load Singularity from the module system and, if you have access to `/exports/igmm/eddie/NextGenResources`, set the Singularity cache directory to the NextGenResources path below. If some containers for your pipeline run are not present, please contact the [IGMM Data Manager](data.manager@igmm.ed.ac.uk) to have them added. You can add these lines to the file `$HOME/.bashrc`, or you can run these commands before you run an nf-core pipeline. +Load Singularity from the module system and, if you have access to `/exports/igmm/eddie/BioinformaticsResources`, set the Singularity cache directory to the BioinformaticsResources path below. If some containers for your pipeline run are not present, please contact the [IGC Data Manager](data.manager@igc.ed.ac.uk) to have them added. You can add these lines to the file `$HOME/.bashrc`, or you can run these commands before you run an nf-core pipeline. -If you do not have access to `/exports/igmm/eddie/NextGenResources`, set the Singularity cache directory to somewhere sensible that is not in your `$HOME` area (which has limited space). It will take time to download all the Singularity containers, but you can use this again. +If you do not have access to `/exports/igmm/eddie/BioinformaticsResources`, set the Singularity cache directory to somewhere sensible that is not in your `$HOME` area (which has limited space). It will take time to download all the Singularity containers, but you can use this again. ```bash module load singularity -export NXF_SINGULARITY_CACHEDIR="/exports/igmm/eddie/NextGenResources/nextflow/singularity" +export NXF_SINGULARITY_CACHEDIR="/exports/igmm/eddie/BioinformaticsResources/nf-core/singularity-images" ``` Singularity will create a directory `.singularity` in your `$HOME` directory on eddie. Space on `$HOME` is very limited, so it is a good idea to create a directory somewhere else with more room and link the locations. @@ -68,7 +68,7 @@ If your eddie terminal disconnects your Nextflow job will stop. You can run Next nohup ./nextflow_run.sh & ``` -### On a wild west node - IGMM only +### On a wild west node - IGC only Wild west nodes on eddie can be accessed via ssh (node2c15, node2c16, node3g22). To run Nextflow on one of these nodes, do it within a [screen session](https://linuxize.com/post/how-to-use-linux-screen/). @@ -92,12 +92,12 @@ screen -r ## Using iGenomes references -A local copy of the iGenomes resource has been made available on the Eddie HPC for those with access to `/exports/igmm/eddie/NextGenResources` so you should be able to run the pipeline against any reference available in the `igenomes.config`. +A local copy of the iGenomes resource has been made available on the Eddie HPC for those with access to `/exports/igmm/eddie/BioinformaticsResources` so you should be able to run the pipeline against any reference available in the `igenomes.config`. You can do this by simply using the `--genome ` parameter. ## Adjusting maximum resources -This config is set for IGMM standard nodes which have 32 cores and 384GB memory. If you are a non-IGMM user, please see the [ECDF specification](https://www.wiki.ed.ac.uk/display/ResearchServices/Memory+Specification) and adjust the `--clusterOptions` flag appropriately, e.g. +This config is set for IGC standard nodes which have 32 cores and 384GB memory. If you are a non-IGC user, please see the [ECDF specification](https://www.wiki.ed.ac.uk/display/ResearchServices/Memory+Specification) and adjust the `--clusterOptions` flag appropriately, e.g. ```bash --clusterOptions "-C mem256GB" --max_memory "256GB" From d9abb7d59fddfdda90d011146a0e82ec10403a51 Mon Sep 17 00:00:00 2001 From: ameynert Date: Wed, 21 Jul 2021 12:23:47 +0100 Subject: [PATCH 3/5] Added eddie config to viralrecon --- pipeline/viralrecon.config | 1 + 1 file changed, 1 insertion(+) diff --git a/pipeline/viralrecon.config b/pipeline/viralrecon.config index 76a15d1..04ec882 100644 --- a/pipeline/viralrecon.config +++ b/pipeline/viralrecon.config @@ -5,3 +5,4 @@ */ includeConfig "${params.custom_config_base}/conf/pipeline/viralrecon/genomes.config" +includeConfig "${params.custom_config_base}/conf/pipeline/viralrecon/eddie.config" From 17fad9dd15008c4da074ec9b042091cd3a6e2253 Mon Sep 17 00:00:00 2001 From: ameynert Date: Wed, 21 Jul 2021 12:27:09 +0100 Subject: [PATCH 4/5] Update pipeline/viralrecon.config Co-authored-by: Harshil Patel --- pipeline/viralrecon.config | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/pipeline/viralrecon.config b/pipeline/viralrecon.config index 04ec882..9300c31 100644 --- a/pipeline/viralrecon.config +++ b/pipeline/viralrecon.config @@ -5,4 +5,7 @@ */ includeConfig "${params.custom_config_base}/conf/pipeline/viralrecon/genomes.config" -includeConfig "${params.custom_config_base}/conf/pipeline/viralrecon/eddie.config" + +profiles { + eddie { includeConfig "${params.custom_config_base}/conf/pipeline/viralrecon/eddie.config" } +} From 21121684200382b360fda18f97b3fd17b37a5e30 Mon Sep 17 00:00:00 2001 From: ameynert Date: Thu, 22 Jul 2021 15:07:19 +0100 Subject: [PATCH 5/5] Set BLASTDB_LMDB_MAP_SIZE This environment variable needs to be set in order to prevent memory allocation errors in the MAKEBLASTDB process. --- conf/pipeline/viralrecon/eddie.config | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/conf/pipeline/viralrecon/eddie.config b/conf/pipeline/viralrecon/eddie.config index 38192ef..8f0463d 100644 --- a/conf/pipeline/viralrecon/eddie.config +++ b/conf/pipeline/viralrecon/eddie.config @@ -1,3 +1,7 @@ +env { + BLASTDB_LMDB_MAP_SIZE=100000000 +} + process { withName : '.*PICARD.*' { clusterOptions = {"-l h_vmem=${(task.memory + 4.GB).bytes/task.cpus}"}