Deseq2 design. Control, siRNA knockdown and scramble siRNA

Question

[marie.lorans]

Hi all, I'm pretty new to R and DGE (consider your self warned) and can't get my head around how to control for transfection effect in my Deseq2 model design. I have three treatments in duplicate samples: Control siRNA target siRNA control/scramble

I made following table:

sampleName                 fileName        condition              transfection
1                          1                ctrl                   0        
2                          2                ctrl                   0
3                          3                siRNA target           1
4                          4                siRNA target           1
5                          5                siRNA control          1
6                          6                siRNA control          1

I tried to create DESeqDataSet as following:

ddsHTSeq <- DESeqDataSetFromHTSeqCount(sampleTable = sampleTable,
                                       directory = directory,
                                       design= ~ condition + transfection)

However, I get the error message:

Error in checkFullRank(modelMatrix) : 
  the model matrix is not full rank, so the model cannot be fit as specified.
  Levels or combinations of levels without any samples have resulted in
  column(s) of zeros in the model matrix.

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

  vignette('DESeq2')

I would really appreciate if anyone could help me and give advice on how to analyse my data so I discard the transfection effect.

Thanks

Answer 1

[i.sudbery]

You can't analys the expriment with this design, because condition=ctrl is the same as transfection=0. You are asking DESeq to assign a particular effect to not being transfected, or being control, but it can't do this, as they are the same thing.

If what you would like to do is compare target knock down to the other two, accounting for the transfection effect, you could try:

sampleName target_KD  transfection
1          0               0
2          0               0
3          1               1
4          1               1
5          0               1
6          0               1

[marie.lorans]

Thanks for the reply. I will try that, thanks :-)