[R] Mantel test for comparing two matrices

Sarah Goslee sarah.goslee at gmail.com
Tue Mar 19 15:01:11 CET 2013


Hi,

A Mantel test doesn't test significant differences between matrices.
It tests for significant correlations between dissimilarities. If your
hypothesis is best expressed in terms of distances, then using the
Mantel test is appropriate.

As for size, 2000 by 2000 is not unmanageable, even if those are raw
data rather than distances already.

Sarah

On Mon, Mar 18, 2013 at 6:23 PM, Houcemeddine Othman <houcemoo at gmail.com> wrote:
> Hi everyone,
> I have two matrices of 2000x2000 size. They consist of a measurement of
> RMSD deviations of a protein structure. The basic question i want
> to answer is the following: are there a significant  differences between
> the two matrices?
> I aim to use the Mantel test. I don't really know if this is a good idea or
> if there are better solutions for these mega matrices.
>
> Thanks
>

--
Sarah Goslee
http://www.functionaldiversity.org



More information about the R-help mailing list