Heatmaps not being generated

[Ashira Lubkin]

Hello,

Lately (since Feb 20), when I have run the GSEA software, the results do not include heatmaps. I used to (last time Jan 17th) get a heatmap for the top pathways that were enriched in each group, which is a really nice way to see quickly which genes most contribute to the enrichment. It's gone now from both the web and file explorer views. 

Thank you,

Ashira

[David Eby]

Hi Ashira,

That seems odd.  Are any of the other plots in the reports missing?  Could you send us a screenshot showing the 'Run GSEA' page with the 'Advanced fields' section fully open?

By any chance, are there special characters (including spaces) in the file name or path of the dataset you are using?  GSEA has known issues with such file names that can cause files to be misplaced; it's best to rename to alphanumeric characters only, with an underscore in place of any special characters.  There can likewise be similar issues if the 'Analysis name' parameter is changed in a similar way.

All of the processing happens on your own computer, so it would come down to possible chissuesanges in the settings, peculiarities of the dataset, or possible local changes on the computer itself since Feb 20.  Depending on what we see in the 'Advanced fields' we might also need to see the dataset that's causing this, which can be sent privately to gsea...@broadinstitute.org (we'll only use it to diagnose issues).

Of course, take a look at any file naming issues first, though.  

Thanks,

David

David Eby
www.gsea-msigdb.org

--
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 visit https://groups.google.com/d/msgid/gsea-help/07804e91-8ee4-437b-aafb-b3ae46804fean%40googlegroups.com.

[Ashira Lubkin]

Hi David,

Thank you for getting back to me.

Looking into it more, it seems that the difference is that the GSEAs I am trying to run now are preranked, but the ones I was doing before were not. 

Here is the run preranked GSEA page with the fields visible, for an analysis I was doing recently:

And here is the page for a GSEA that I ran successfully in January. I re-ran it again, and it did indeed generate the heatmaps. 

Is there a way to get heatmaps for a preranked GSEA?

Thank you for your help,

Ashira

You received this message because you are subscribed to a topic in the Google Groups "gsea-help" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/gsea-help/0-RL4b5Isw4/unsubscribe.
To unsubscribe from this group and all its topics, send an email to gsea-help+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/gsea-help/CAFEo9XhwproHk4WVNQe2GaUgnX8rmqU78DisK0g-Bnv90qcYwQ%40mail.gmail.com.

[David Eby]

Ah, I see!  Yes, that explains the difference.

If you check the heatmap from one of your former experiments, you'll see that it uses the phenotype labels (from your CLS file) as the column headers.  That is, the heatmap is showing the difference in gene expression across the samples.  Since the Preranked function omits any phenotypic information, there's no way to create such a heatmap in the context of this type of analysis.

The best advice I could give in this case is to try to produce these heatmaps - with something like R or Python or even Excel - from the upstream data at a point where the samples are still distinct from one another.  It doesn't really solve your problem exactly, however, since the GSEA heatmaps show only the genes that map to the given gene set.  You would also need to "slice" just those rows out of the dataset as well.

For reference, this particular heatmap uses the input dataset directly as given (unmodified) and generates the heatmap for just those matching rows.

Regards,

David

David Eby
www.gsea-msigdb.org

To view this discussion visit https://groups.google.com/d/msgid/gsea-help/CAPnk1TLdnCFd3OMcRgM%2BAbSPyTyMbOnGa__sc-JF7fU-1kpmXA%40mail.gmail.com.

[Ashira Lubkin]

Oh, yeah, that makes sense. I haven't given it the data to make the heat maps. Thank you for explaining. 

Best, 

Ashira 

To view this discussion visit https://groups.google.com/d/msgid/gsea-help/CAFEo9XjJxr1tzVH0vLP5qnZoyYHb9yqkeexdMxeyxFkkL7rUHg%40mail.gmail.com.