Question

what happened in limma or Desq2 for ANOVA

0

Entering edit mode

Bioinformatics ▴ 30

@bioinformatics-10931

Last seen 2.3 years ago

United States

I know that for limma we use the following for when we have 2 control and 2 treated

design <- model.matrix(~c(rep(0,2),rep(1,2)))

and for Deseq2

condition <- factor(c("Control","Control","Treated","Treated"))

Now I have this questions one specifically for limma

1- does it matter in limma which design I use? for example if I have the frirst two column as control , should I assign them to 0 or 1?

2- what happens after enrichment analysis (how does the Deseq2 different from limma)?

limma • 781 views

ADD COMMENT • link updated 6.2 years ago by Gordon Smyth 50k • written 6.2 years ago by Bioinformatics ▴ 30

Gordon Smyth · Answer 1 · 2018-01-25

0

Entering edit mode

Gordon Smyth 50k

@gordon-smyth

Last seen 19 minutes ago

WEHI, Melbourne, Australia

Why not use

design <- model.matrix(~condition)

It's simple and easy.

The documentation would tell you this, especially Section 9.2 of the limma User's Guide. What don't you avail yourself of it?

ADD COMMENT • link 6.2 years ago Gordon Smyth 50k

0

Entering edit mode

actually one ignores the intercept and one involve the intercept . I am wondering which way is the right way to do ?

ADD REPLY • link updated 6.2 years ago by Gordon Smyth 50k • written 6.2 years ago by Bioinformatics ▴ 30

0

Entering edit mode

All the approaches described in Section 9.2 are correct and none of them "ignores" the intercept.

I'm not going to answer any more questions from you until you show some willingness to read the documentation. It would also be appreciated if you made a bit more effort to ask a better-formed question. At the moment, you don't seem to have an actual problem.

ADD REPLY • link 6.2 years ago Gordon Smyth 50k