search

RitchieLabIGH / IRFinder

MIT License
13 stars 10 forks source link

Problem with IRFinderDiff #21

Closed JYuyangLu closed 1 year ago

JYuyangLu commented 1 year ago

Dear Claudio,

I met a problem with IRFinderDiff. 

converting counts to integer mode
using pre-existing size factors
estimating dispersions
Error in checkForExperimentalReplicates(object, modelMatrix) : 

  The design matrix has the same number of samples and coefficients to fit,
  so estimation of dispersion is not possible. Treating samples
  as replicates was deprecated in v1.20 and no longer supported since v1.22.

Calls: DESeq ... estimateDispersions -> .local -> checkForExperimentalReplicates
Execution halted

Does it relate to the R package version or something else?

Best, Yuyang

CloXD commented 1 year ago

Hello Yuyang, the error is given by DESeq2 when you try to use only one replicate for each sample, is it the case? Could you give me more information about your design? Cheers, Claudio

JYuyangLu commented 1 year ago

Hi Claudio,

Thank you for your quick response. I see. I included all the replicates I have, and now the program runs perfectly. BTW, I saw some boxplots in the output folder. I would like to confirm that are they as a subset of IR regions with significant p values? Thank you!

Yuyang

CloXD commented 1 year ago

Hello Yuyang, yes, exactly, those are the IRratios of the introns having adjusted p-value lower than 0.05. Cheers, Claudio