I was wondering whether using the partial coefficient of determination is appropriate for variance decomposition, when using DESeq2? In other words, I want to calculate the proportion of variance (in the observed data; i.e., read counts) explained by each covariate in the model.
I basically want to know the following
(1) is it a good and meaningful measure or not, considering we're dealing with a glm?
(2) is there a way I can achieve equivalent information by handling "deviances" instead?
Any comment or hint would be appreciated. Also, I'd appreciate if anyone could suggest a good reference book or paper on this.
Thanks a bunch!
I basically want to know the following
(1) is it a good and meaningful measure or not, considering we're dealing with a glm?
(2) is there a way I can achieve equivalent information by handling "deviances" instead?
Any comment or hint would be appreciated. Also, I'd appreciate if anyone could suggest a good reference book or paper on this.
Thanks a bunch!
Comment