[adegenet-forum] Confirming the use of linked SNPs for sPCA

[Alastair Potts]

Hi Thibaut,
You suggested that I use sPCA for my linked SNP data. The results are 
pretty cool as they fit in with my study system very nicely. I just 
wanted to confirm that linked SNPs can be used for sPCA. I just thought 
I should query this as your Jombart (2009; Heredity) paper suggests its 
use is confined to microsats.

I have converted the SNP data in the following way for adegent
From:
      1 2 3
s1 A Y A
s2 R C T
s3 G T T

To:
      a1.a a1.g a2.c a2.t a3.a a3.t
s1  1      0      0.5  0.5    1     0
s2  0.5   0.5     1    0      0     1
s3  0      1        0    1      0     1

and then converted this to a genind object.
If I can use SNPs for sPCA (and I can't really think of a reason why I 
can't) is the above conversion correct for analysis in adegenet?

Thanks very much for your time,
Cheers
Alastair