I was trying Next maSigPro on some RNA-seq data and while using p.vector(data,d,....):

Error in if (model.glm$null.deviance == 0) { :
  missing value where TRUE/FALSE needed

This occurred because the gene in question had 0 normalized expression value for every col.
Should this occur even when setting the min.obs>0 ? Because the default is 3 (min.obs=3), then I tried to set it manually and the error persists.
I can filter it manually, but maybe the package could treat this once it has an option for doing so.

Hi,

As you say, the error is due to there are a lot of ceros in a row (for a gene).

min.obs is a parameter that controls the missing data but it doesn't control the cero values. Actually maSigPro doesn't have any argument to control de ceros. I think it is a good idea to consider this in p.vector function to improve the proccess and make it easier.

Thank you,

María Nueda.