R aborted during WGCNA under Mac OS environment

[chubin]

Hi all,
I am encountering a continuous session abortion problem with my RStudio. I am using R version 4.1.3 and RStudio 2022.02.1+461 "Prairie Trillium" Release (8aaa5d470dd82d615130dbf663ace5c7992d48e3, 2022-03-17) for macOS
Mozilla/5.0 (Macintosh; Intel Mac OS X 12_2_1) AppleWebKit/537.36 (KHTML, like Gecko) QtWebEngine/5.12.10 Chrome/69.0.3497.128 Safari/537.36. on my Macbook.

While running WGCNA package with my data and everything went pretty well till yesterday.I ran the same script with similar dataset last month and everything went smoothy. I tried to reinstall both R and Rstudio but it keeps showing the same fetal error.

Below is where I encounter the abortion:

        softPower = 10;
        adjacencies = array(0, dim = c(nSets, nGenes, nGenes));

It would be appreciated if anyone could help me solve this issue.

Many thanks!

[nirgrahamuk]

I don't doubt that your session crashed but I am skeptical that it did so on either of the two statements you identified. These two seem to only assign some simple R objects.
perhaps it crashes after, when they are used somewhere ?
Regardless, in your shoes, given that this WGCNA is a package which likely has high performance code (according to CRAN) CRAN - Package WGCNA (r-project.org) RCPP and various parallelisation libs, it would probably be best to alert the package author/maintainer.

[chubin]

That's also why I have no idea how to modify my script because my gut feeling is the following statements would be more likely to cause crushes (like blockwise module building or pairwise comparison). However, since yesterday, I've ever go beyond the adjacencies step.

[nirgrahamuk]

what are the values of nSets and nGenes at this point ? are they just some integer values ?

[chubin]

Yes, my data is basically sequence read counts and it's just some integers. nSets=2 and nGenes=143619.

[nirgrahamuk]

ah well, thats your problem

adjacencies = array(0, dim = c(2, 143619, 143619));
Error: cannot allocate vector of size 307.4 Gb

Perhaps you could achieve that with some cloud compute...

[chubin]

wow, that's crazy. Now I wonder how I got my result last month.
Thanks for the help.