Cholmod error 'problem too large' at file ../Core/cholmod_dense.c, line 105

[ksr2018]

Hi @farrellja ,
I'm trying URD on a dataset with about 50k cells and when I try to run:

myURDobject.biased.tm <- as.matrix(pseudotimeWeightTransitionMatrix(myURDobject, "pseudotime", logistic.params=myURDobject.ptlogistic))

it gives the following error:

Error in asMethod(object) :
Cholmod error 'problem too large' at file ../Core/cholmod_dense.c, line 105

I realize this may be related to issue #6, but I do not know how to fix it. Can you please help?

Thank you.

[farrellja]

Hi @ksr2018,

I have updated the pseudotimeWeightTransitionMatrix function, which was having an error above 46k cells. To avoid it, it divides the data into pieces, processes them separately, and then reassembles it. Try installing from the debug branch (devtools::install_github(repo="farrellja/URD@debug"), should be version 1.0.3.9002) and try again. You probably will also want to remove the as.matrix() in the above call to keep the output as a sparseMatrix.