Hello,
I have performed a two-factor design RNA seq experiment (treatment and sex) and would like to know if these factors have a significant effect on gene expression.
To do this I followed the instructions as in the DESeq manual and fitted the following models:
fit0 <- fitNbinomGLMs( cds, count ~ treatment)
fit1 <- fitNbinomGLMs( cds, count ~ treatment + sex)
fit2 <- fitNbinomGLMs( cds, count ~ treatment + sex + treatment * sex)
Is it possible to determine which of these models fits the data best?
Typically when fitting glms I would use AIC to determine model term significance (i.e. does fitting an interaction term significantly improve the fit of the model?). Is there any way to do this is DESeq?
Many thanks,
Darren
I have performed a two-factor design RNA seq experiment (treatment and sex) and would like to know if these factors have a significant effect on gene expression.
To do this I followed the instructions as in the DESeq manual and fitted the following models:
fit0 <- fitNbinomGLMs( cds, count ~ treatment)
fit1 <- fitNbinomGLMs( cds, count ~ treatment + sex)
fit2 <- fitNbinomGLMs( cds, count ~ treatment + sex + treatment * sex)
Is it possible to determine which of these models fits the data best?
Typically when fitting glms I would use AIC to determine model term significance (i.e. does fitting an interaction term significantly improve the fit of the model?). Is there any way to do this is DESeq?
Many thanks,
Darren
Comment