EBSeqHMM error with many replicates per time point

Question

[Akos]

Hi All,

Anyone using EBSeqHMM out there?

I would like to use EBSeqHMM on a time - series RNA-seq data, where I have 5 time points and more than 10 replicates for each time point. However, the time points have different number of replicates.

First, when I tested EBSeqHMM on a dummy dataset with only triplicates per time-point it was working. However, with my real data It gives error message at the following step:

EBSeqHMMGeneOut <- EBSeqHMMTest(Data=matrix2, sizeFactors=Sizes, Conditions=Conditions, UpdateRd=5)

Error in if (max(Pi.up) > PIBound) { : 
    Missing value where TRUE/FALSE needed

Anyone knows what might be the problem?

Or is there a way to use EBSeqHMM with many and uneven replicates per time points? Thanks in advance.