degPatterns crashes because of stack overloading about degreport HOT 7 CLOSED

Hi,
Thank you for the debugging. I will make a note and try to do this in the next two weeks, if you find time for a PR, I will merge as well.
Just to make sure, is this line what you found to be the issue?

from degreport.

Hi,
I am facing the same problem with degPatterns() with the number of genes input ranging from 1000-4000. I've used this function multiple times in the past without a glitch for more than 6000 genes, generating the object, cluster image and, exporting the group-wise gene list.
Temporarily I've been managing by running the following code at bash before opening R from the same terminal ulimit -s unlimited. With this, the object gets created and I can export the group-wise gene lists but I cannot save the cluster image since it gets replaced by the dendrogram. Has this issue been resolved yet or is there a way to get around this?
Thanks so much!

from degreport.

Hi, I am still trying to figure out the issue. It is hard because I cannot reproduce in MACOSX. I will try to remove the dendrogram code and see if that helps.

from degreport.

I have pushed an temporary fix to avoid using dendextend in the main branch of this repo. Could you test loading the code from this branch and see if the errors go away?

from degreport.

Hello,

I am currently on R 4.3.2, DEGReport v 1.38.5, and macOS 14.4 and I am still having this issue. As noted by previous commentators, the error appears when a large number of genes are submitted for clustering (in my case, I would get this error "C stack usage 7969268 is too close to the limit" when the number of genes was over 5000). I just reduced the number of genes to 4300 by using a more stringent p-value, and I was able to successfully cluster them. However, it would be cool if we could cluster without having to make alterations to our thresholds.

Thanks for looking into this!

from degreport.

Can you install version: 1.39.6? Right now it is GitHub main, so you will need to install it with devtools? If that works I will merge it with bioconductor. thanks!

from degreport.

Hi!

I just tried out v 1.39.6 and had no issues clustering a large number of genes. Thank you for the quick fix!

from degreport.