Table 2.

10 times repeated 5-fold cross-validated F1 score for 19 population classes from 1000 Genomes Project using SVM, PCA or GTM

Ancestry	1000G code	Population	PCA 8-NN	SVM 10 PCs	GTM 3 PCs	GTM 10 PCs
EAS	CHB	Han Chinese	0.20±0.01	0.78±0.01	0.45±0.04	0.75±0.01
EAS	JPT	Japanese	0.37±0.02	1.00±0.00	0.80±0.01	1.00±0.00
EAS	CHS	Southern Han Chinese	0.34±0.02	0.80±0.01	0.54±0.02	0.80±0.01
EAS	CDX	Chinese Dai	0.24±0.02	0.10±0.02	0.51±0.03	0.44±0.08
EAS	KHV	Kinh in Vietnam	0.44±0.01	0.68±0.00	0.63±0.01	0.71±0.01
EUR	CEU+GBR	Northern/Western Eur.	0.75±0.01	0.99±0.00	0.79±0.01	0.99±0.00
EUR	TSI	Toscani	0.46±0.01	0.74±0.02	0.58±0.01	0.54±0.06
EUR	FIN	Finnish	0.95±0.01	0.99±0.00	0.91±0.01	0.99±0.01
EUR	IBS	Iberian	0.35±0.03	0.81±0.01	0.35±0.04	0.74±0.02
AFR	YRI	Yoruba in Nigeria	0.30±0.02	0.69±0.00	0.15±0.03	0.66±0.03
AFR	LWK	Luhya	0.67±0.01	1.00±0.00	0.59±0.01	1.00±0.00
AFR	GWD	Gambian	0.26±0.02	0.94±0.02	0.23±0.02	0.78±0.07
AFR	MSL	Mende	0.25±0.03	0.93±0.02	0.35±0.03	0.81±0.04
AFR	ESN	Esan in Nigeria	0.28±0.02	0.00±0.01	0.19±0.05	0.28±0.13
AMR	PUR	Puerto Ricans	0.90±0.01	0.86±0.02	0.90±0.01	0.87±0.03
AMR	CLM	Colombians	0.69±0.01	0.85±0.01	0.84±0.01	0.82±0.02
AMR	PEL	Peruvians	0.88±0.01	0.97±0.01	0.94±0.01	0.95±0.01
SAS	PJL	Punjabi	0.89±0.01	0.96±0.01	0.96±0.01	0.96±0.00
SAS	BEB	Bengali	0.95±0.01	0.96±0.01	0.96±0.01	0.96±0.01
Overall			0.54±0.00	0.80±0.00	0.61±0.01	0.80±0.01

SVM 10 PCs = support vector machine classification model using 10 principal components, PCA 8-NN = k-nearest neighbours model based on 2D PCA map (k = 8), GTM 3 or 10 PCs = bayesian classification model based on generative topographic mapping using 3 or 10 principal components. Ancestry codes: EAS = East Asians, EUR = Europeans, AFR = Africans, AMR = Admixed Americans, SAS = South Asians. CEU and GBR were merged into one class. Each value is an average with 95% confidence interval