How to fix DESeq2 Error in checkFullRank(modelMatrix)

Question

[marco.barr]

Hi, this is my colData

from my DESeq2 analysis returns this error:

ddsTxi <- DESeqDataSetFromTximport(txi, colData = colData, design = ~concentration+condition)      
Errore in checkFullRank(modelMatrix) : 
  the model matrix is not full rank, so the model cannot be fit as specified.
  One or more variables or interaction terms in the design formula are linear
  combinations of the others and must be removed.

Please read the vignette section 'Model matrix not full rank':

vignette('DESeq2')

How can I reshape my data to avoid linear combinations? Thanks

Answer 1

[i.sudbery]

Your conecntration and condition are effectively the same thing, there is no way this can be rearranged to to fit effects for both concentration and condition. What are you trying to achieve?

[marco.barr]

In the concentration column there are 3 different treatments with their relative concentrations. I want to see if the differential expression changes between those treated at 1x concentration and those treated at 10x. I wanted to set up my experimental design with the two covariates of both treatment and concentration. How can I do?

[ATpoint]

Treatment and concentration is the same. 10x is always B, 1x is always A, untreated is always 0. Use one or the other column, but not both for the analysis.

[i.sudbery]

If you want to detect the difference that the concentration of the treatment makes to gene expression, then you just want ~concentration not ~concentration + treatment