Search
Question: Error in checkFullRank(modelMatrix) : the model matrix is not full rank, ?
0
8 months ago by
Björn0
CH
Björn0 wrote:

I used following command to filter out 3 groups (NC, BF, CF) out of 5 groups while remaining two groups are "AF" and "DF".

design_NCaF <- filter(design, Groups1!="NC",Groups1!="BF",Groups1!="CF")

However, at following command

NCaF <- DESeqDataSetFromMatrix(data_max_woNCaF, colData=design_NCaF, design = ~Groups1 + Patient)

It gives me following error message

factor levels were dropped which had no samples
Error 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.

Is it because of remaining two groups have different sample size (AF=6 and DF=3) ?

modified 7 months ago by Michael Love18k • written 8 months ago by Björn0
0
8 months ago by
Michael Love18k
United States
Michael Love18k wrote:

You need to droplevels() from the factor after filtering, e.g. if you filter out samples from 'group':

dds$group <- droplevels(dds$group)