Closed gskoufos closed 2 years ago
Hi @gskoufos,
Thank you very much for your interest in our package and thank you for providing such a detailed description on the procedures! Based on your description and the output from iDEA, everything looks good to me. So can you provide me the following information or try these:
1)What is your operating system? If your system is Windows, that might be related to the issue that windows does not support mclapply (at least at the time when we implemented the package), a function related to the parallel computing. For this issue, I need to take time to find a replacement for this function.
2)Then perhaps, try to subset 5 gene sets, for example, when you create iDEA object, you can write:
idea <- CreateiDEAObject(summary, mouseGeneSets[,1:5], max_var_beta = 100, min_precent_annot = 0.0025, num_core = 10)
Pleas note that when you subset 5 gene sets, make sure that these gene sets are not filtered (you can check the iDEA input summary ## number of annotations should be > 0). If they are filtered out, just use another 5 or 10 gene sets This just helps me to see whether it still stuck at 0%.
3) Also, just in case, how many genes are DE genes based on your original MAST output, i.e. p.adjust < 0.05?
Hi,
I'm trying to use iDEA in my scRNA-Seq mouse datasets using your tutorial but I'm facing an issue.
I got my (cluster-specific) DE (and non-DE) genes using MAST on a Seurat object and calculated summary statistics. This is how the dataframe holding summary statistics looks like this:
I'm also using the mouseGeneSets you provide through the function
data(mouseGeneSets)
. Here is the output of the commandmouseGeneSets[1:3,1:3]
:As you can see, I'm using gene names (instead of IDs) in both the mouseGeneSets and MAST results.
Subsequently, I'm creating an iDEA object using the following command:
idea <- CreateiDEAObject(summary, mouseGeneSets, max_var_beta = 100, min_precent_annot = 0.0025, num_core = 10)
After a few seconds, the iDEA object is ready.
The format of the idea@summary table looks like this:
Finally, I'm executing the following command:
idea <- iDEA.fit(idea, fit_noGS = FALSE, init_beta = NULL, init_tau = c(-2, 0.5), min_degene = 5, em_iter = 15, mcmc_iter = 1000, fit.tol = 1e-5, modelVariant = FALSE, verbose = TRUE)
and the message I get in the console is the following:
Then, I waited for a long time (more than an hour) but nothing happened. The progress is stuck at 0%.
Any idea what's causing this?
Thank you for your time!