We use STRUCTURE, which is very similar to ADMIXTURE, but we have not tried to extract the SNPs driving the ancestry components because to date we have preferred to look at the total result. Nonetheless, it should be a simple part of the ancestry output where the contribution of each SNP is given.

this paper may help you:

Ralph E. McGinnis, Visualizing Chromosome Mosaicism and Detecting Ethnic Outliers by the Method of “Rare” Heterozygotes and Homozygotes (RHH) , Human Molecular Genetics, 2010, Vol. 19, No. 13 2539–2553, doi:10.1093/hmg/ddq102

I'm not an expert on the field, but I saw it referred on an interesting blog entry where the author presented a worked example to detect in fact "ethnic outliers".