LIMMA one-class comparison p-value missing

Help
Markus
2013-07-04
2013-07-04
  • Markus
    Markus
    2013-07-04

    Hi,

    first of all I would like to thank the TMEV developer team for making this great piece of software available!

    Now to my problem:
    I am trying to analyze log transformed ratio data from five replicates using a LIMMA one-class comparison. I assume that this would be technically similar to a one-class t-test against a defined value (0, as in my case). Am I right? When I calculate the LIMMA and plot the data, it seems to make sense. My problem is that no p-value or other statistics are returned in the table, except group mean, std and an F value that is 0 in all cases. Without these data the computation of significant data points is unfortunately worthless to me. Am I doing something fundamentally wrong, or is this a bug? I noticed that a two-class comparison works fine, returning all the computed statistics.

    Markus