Human ancestry indentification under resource constraints -- what can one chromosome tell us about human biogeographical ancestry?

Table 5 Comparative Performance In Sub-Population-level Ancestry classification

Pairwise sub-populations	Continent	Method	Data size	Datasets	Classification rate (%)	Number of attributes used
CEU-TSI	europe	ethnopred [38]	267	hapmap iii	86.6 ± 2.4	180 SNPs
CHB-JPT	east asia	ethnopred [38]	250	hapmap iii	95.6 ± 3.9	877 SNPs
LWK-MKK	africa	ethnopred [38]	294	hapmap iii	95.9 ± 1.5	341 SNPs
JPT-CHB	east asia	bayesian [39]	9104	own collection	74.9 (77.2***)	15 STR loci
JPT-KOR	east asia	bayesian [39]	731	own collection	67.9 (63.7)	15 STR loci
CHB-KOR	east asia	bayesian [39]	731	own collection	69.6 (62.4)	15 STR loci
–	europe	proposed	503	1000 genome phase 3	76.6*	58 snps**
–	africa	proposed	661	1000 genome phase 3	87.02*	87 snps**
–	east asia	proposed	504	1000 genome phase 3	73.3*	68 snps**

*Average accuracy of all pairwise sub-population classifications within the given continent
**Average number of SNPs required in all pairwise sub-population classifications within the given continent
***Results obtained without normalization

ISSN: 1755-8794