Fatal error: /atac_dnase_pipelines/modules/conf.bds, line 39. Java heap space

[nteran]

I am attempting to run bds atac using SCG.

The listed fatal error is java heap space, but changing the memory (anywhere from 30G - 900G) gives the same error and the -use_sambamba_markdup flag does not change anything.

The last line in the log is "Creating checkpoint file: Config or command line option disabled checkpoint file creation, nothing done."

My command is: ~/montgomery/atac/scripts/atac_dnase_pipelines/utils/bds_scr RD002 ~/montgomery/atac/scripts/atac_dnase_pipelines/atac.bds -nth 12 -mem_dedup 30G -mem_ataqc 30G -use_sambamba_markdup -fastq1_1 /labs/smontgom/nteran/motrpac/atac/tissue_tests/run6/merged/RD002_S2_R1_001.fastq.gz -fastq1_2 /labs/smontgom/nteran/motrpac/atac/tissue_tests/run6/merged/RD002_S2_R2_001.fastq.gz -species_file /srv/gsfs0/projects/montgomery/mdegorte/Africa/ATAC/atac_dnase_pipelines/myGenomes/rn6/bowtie2_index/rn6.fa -system slurm -q_for_slurm_account -q smontgom

The log file is attached/ RD002.BDS.log

[leepc12]

https://github.com/kundajelab/atac_dnase_pipelines#bigdatascript

Did you add Java options to your ~/.bashrc?

[nteran]

I had not.

I added it and reran it. It dies in the same place, although it does note that it picked up _JAVA_OPTIONS.

On Wed, May 16, 2018 at 10:40 AM, Jin Lee [email protected] wrote:

https://github.com/kundajelab/atac_dnase_pipelines#bigdatascript

Did you add Java options to your ~/.bashrc?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/kundajelab/atac_dnase_pipelines/issues/115#issuecomment-389604753, or mute the thread https://github.com/notifications/unsubscribe-auth/ANGl8uR-z-9OogfK8y-0sPyFM0a25GRfks5tzGSLgaJpZM4UBuXe .

[leepc12]

You didn't specify -species and -species_file correctly. Please see README for details.

Your species_file is wrong. It's pointing to a fasta file. It should be a text file (INI) containing genome data file location.

-species_file /srv/gsfs0/projects/montgomery/mdegorte/Africa/ATAC/atac_dnase_pipelines/myGenomes/rn6/bowtie2_index/rn6.fa

[nteran]

Thank you for your help.

I think I'm supposed to be passing these variables in a config file, but it seems to work when I add them as direct flags. However, it now seems to be spinning its wheels at the alignment stage.

01:28:38.892 ExecutionerClusterGeneric 'ClusterGeneric[31]': Tasks [ClusterGeneric[31]] Pending: 0 Running: 1 Done: 1 Failed: 0 | PID | Task state | Task name | Dependencies | Task definition

         |
            | ------- | ----------------- | --------------- |

------------ |

| | 1610998 | running (RUNNING) | bowtie2_PE rep1 | | bowtie2 -X2000 --mm --local --threads 12 -x /srv/gsfs0/projects/montgomery/mdegorte/Africa/ATAC/atac_dnase_pipelines/myGenomes/rn6/bowtie2_index/rn6.fa; -1 /labs/smontgom/nteran/motrpac/atac/tissue_tests/run6/merged/RD002_S2_R1_001.fastq.gz -2 /labs/smontgom/nteran/motrpac/atac/tissue_tests/run6/ |

On Wed, May 16, 2018 at 12:06 PM, Jin Lee [email protected] wrote:

You didn't specify -species and -species_file correctly. Please see README for details.

Your species_file is wrong. It's pointing to a fasta file. It should be a text file (INI) containing genome data file location.

-species_file /srv/gsfs0/projects/montgomery/mdegorte/Africa/ ATAC/atac_dnase_pipelines/myGenomes/rn6/bowtie2_index/rn6.fa

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/kundajelab/atac_dnase_pipelines/issues/115#issuecomment-389631323, or mute the thread https://github.com/notifications/unsubscribe-auth/ANGl8lNTegvc5sqjRovww_W68Tu64L7Dks5tzHifgaJpZM4UBuXe .