Genome-wide scan or only SNPs of interests

[Yiguan]

Hi Aaron,

Thanks for providing such an amazing tool. I basically have two questions when applying clues to my real genome data.

• which SNPs in under selection?

• what's the selection coefficicents for the SNPs unders selection?

To answer these questions, does that mean I need to scan each SNP along the whole genome? Then for each SNP, I could have the scan results, e.g:

Loading data and initializing model...
(100, 1) (100, 21)
Optimizing likelihood surface using Nelder-Mead...
        (Importance sampling with M = 100 Relate samples)

##########

logLR: 0.2364

MLE:
========
epoch   selection
0-1000  0.00254
0

And then logLR: 0.2364 indicates the likelihood of the SNP under selection? and 0.00254 being the selection coefficient?

If so, then I think the biggest challenge is compuation if all the SNPs are scanned individually, which might take many weeks...

In short for my question, is 'clues' suitable for genome-wide scan without any knowledge of candidate SNPs?

Thanks very much!

Yiguan