Hi,

I do have samples collected from 2 different region and from each region we collected sample from 9 villages like this

Rajshahi ----9 villages Mymensingh --9 villages

Now I want to do differential microbial abundance between Rajshahi and Mymensingh, but by controlling for the effect of villages:

when I pass below code it show error:

> psfdds = phyloseq_to_deseq2(ps5, ~ Villages + Region)
converting counts to integer mode
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.

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

But when I pass below code there is no error

> psfdds = phyloseq_to_deseq2(ps5, ~Region)
converting counts to integer mode

and sample data looks like this:

> sample_data(ps5)
Sample Data:        [48 samples by 5 sample variables]:
                       X.SampleID            Treatment Compartment   Villages              Region           
Root-1.S35.L001   Root-1.S35.L001                  ash        Root  Kholabari            Mymensingh
Root-10.S75.L001 Root-10.S75.L001                  ash        Root     Matura            Rajshahi
Root-13.S5.L001   Root-13.S5.L001                  ash        Root      Gokul            Rajshahi
Root-14.S16.L001 Root-14.S16.L001                  ash        Root       Jiol            Mymensingh
Root-17.S26.L001 Root-17.S26.L001                  ash        Root    Chandia            Rajshahi

could you please advise me how I can see the effect of region by controlling the effect of villages?

thanks nabiyogesh

If each village is in exactly one region, you can't put village and region in the design. Pick one or the other.

Did you take a look at the vignette section that the error message points you to? It has useful information about why various designs are not possible. Take a look there first please and come back with specific questions.