DESeq2 Not full Rank - Age & Medication

0

Entering edit mode

2.6 years ago

Anand ▴ 40

Hi. I've a data matrix which looks like this:

Metadata

I'm trying to carrying out DGE analysis controlling for age and medication so I used the following:

design(ddsMF)<-formula(~age+medication+group)
ddsMF<-DESeq(ddsMF)

And end up with the error, "Full model matrix is less than full rank."

I saw the vignettes on the DESeq2 tutorial but am not sure if they would work here given the large age range. Any advice would be most welcome.

medication DESeq2 age rank • 1.6k views

ADD COMMENT • link updated 2.6 years ago by Devon Ryan 104k • written 2.6 years ago by Anand ▴ 40

2

Entering edit mode

Odds are good that you have a typo somewhere in your design matrix. Make sure group and medication are factors with only 2 levels and that age is numeric, rather than a factor. Otherwise the model matrix should be full rank.

ADD REPLY • link 2.6 years ago by Devon Ryan 104k

0

Entering edit mode

Fixed! Thanks Devon.

ADD REPLY • link 2.6 years ago by Anand ▴ 40

0

Entering edit mode

For future people with the same issue, can you mention exactly what was wrong?

ADD REPLY • link 2.6 years ago by Devon Ryan 104k

Login before adding your answer.