Ancestry inference using principal component analysis and spatial analysis: a distance-based analysis to account for population substructure

Table 1 Comparison among 7 subpopulations within Europe using Hotelling’s T² test

Population1	Population2	Statistic	P-value	P-value*
N. European	S. European	334.97	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴
N. European	Great Britain	331.63	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴
N. European	Russian	148.56	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴
N. European	Arab	81.87	1.12 × 10⁻¹⁴	< 1 × 10⁻⁴
N. European	Basque	181.06	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴
N. European	Jews	362.28	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴
S. European	Great Britain	680.60	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴
S. European	Russian	713.40	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴
S. European	Arab	334.90	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴
S. European	Basque	710.36	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴
S. European	Jews	1108.18	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴
Great Britain	Russian	865.25	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴
Great Britain	Arab	646.45	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴
Great Britain	Basque	1165.79	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴
Great Britain	Jews	73.14	7.77 × 10⁻¹⁵	< 1 × 10⁻⁴
Russian	Arab	17.64	1.04 × 10⁻⁸	1 × 10⁻⁴
Russian	Basque	4.96	0.0014	0.0014
Russian	Jews	1436.50	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴
Arab	Basque	16.82	2.34 × 10⁻⁸	< 1 × 10⁻⁴
Arab	Jews	1038.41	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴
Basque	Jews	1366.32	< 1 × 10⁻¹⁶	< 1 × 10⁻⁴

P-value* is computed using permutation test which estimates the non-parametric P-value for the hypothesis test in Hotelling’s T² test

ISSN: 1471-2164