Question: estimateDispersions fails even with fitType local
0
4.0 years ago by
United States
sahiilseth0 wrote:

Hi DEXSeq Experts,

I have a dataset with 17 tumor normal pairs, and am trying to explore differential splicing. My first attempt at estimateDispersions failed, so I tried again with fitType="local", now this fails with the following error:

design = ~sample + exon + condition:exon,

> dxd <- estimateDispersions(dxd, fitType="local", BPPARAM = MulticoreParam(18))
using supplied model matrix
Error in .rbind.SummarizedExperiment(args) :
'...' objects must have the same number of samples

modified 4.0 years ago by Michael Love26k • written 4.0 years ago by sahiilseth0
Answer: estimateDispersions fails even with fitType local
0
4.0 years ago by
Michael Love26k
United States
Michael Love26k wrote:

It might help if you provide the error message for the first attempt as well.

It may have something to do with the number of cores, as rbind is used to collect results after sending to the workers.