How to get enrichment plot in SVG format that is located at around 1000th position?

Dong

unread,

Sep 12, 2021, 11:05:06 AM9/12/21

to gsea-help

Hi,

I am running my data against C7 and GO.BP dataset collections that contain around 10K genesets. I found that some of enriched genesets that I am interested in are located at around 1000th position. To get the enrichment plots that I am interested, I have to assign big number to the "Plot graphs for the top sets of each phenotype". Meanwhile, to make figures at high resolution in illustrator, I also have to choose to create SVG plot images. As a result, the running ate a lot of ram and also took a long time. In most cases, the running failed. Do you have suggestion on this?

Thanks.

Dong

Anthony Castanza

unread,

Sep 12, 2021, 2:27:18 PM9/12/21

to gsea-help

You can try running the set individually, you wouldn't be able to use the statistics from the individual run and the NES might vary slightly but the raw enrichment score should be the same as should the plot.

-Anthony

Anthony S. Castanza, PhD
Curator, Molecular Signatures Database
Mesirov Lab, Department of Medicine
University of California, San Diego

--
You received this message because you are subscribed to the Google Groups "gsea-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gsea-help+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gsea-help/fab864d0-b928-42e0-8c9d-1594c2eca12fn%40googlegroups.com.

Dong

unread,

Sep 12, 2021, 3:53:57 PM9/12/21

to gsea-help

So, you meant that, if I run the set individually, I will get the same enrichment plot but different statistics and NES? And I will be allowed to combine the plot from individual run and statistic and NES from batch run to make final figure?

Thanks.

Dong

Anthony Castanza

unread,

Sep 12, 2021, 4:36:46 PM9/12/21

to gsea-help

Yes, the enrichment plot is based on only on the input data itself and doesn't use any information from the permutation testing or global statistics, so it shouldn't be an issue to use the plot from the individual run but the statistics from the "full" run.

-Anthony

Anthony S. Castanza, PhD
Curator, Molecular Signatures Database
Mesirov Lab, Department of Medicine
University of California, San Diego

To view this discussion on the web visit https://groups.google.com/d/msgid/gsea-help/c5509ef3-79c7-4c65-bced-c61bcf5e9effn%40googlegroups.com.

Dong

unread,

Sep 12, 2021, 5:33:39 PM9/12/21

to gsea-help

Hi Anthony,

Thank you so much!

Dong

Reply all

Reply to author

Forward