An Eigenvalue test for spatial principal component analysis

Montano, V.; Jombart, T.

doi:10.1186/s12859-017-1988-y

BMC Bioinformatics

Table 1 Significant results for global test (g test), local tests (l test), and spca_randtest (r test +/−) for random, global and local patterns using 200 loci per individual. IS, SS, IBD indicate the migration models (see Methods); different migration rates are coded by number: 1 = 0.005, 2 = 0.01 and 3 = 0.1

From: An Eigenvalue test for spatial principal component analysis

200 SNPs		Random Patterns				Global Patterns				Local Patterns
Models	Significance level	g test	r test (+)	l test	r test (−)	g test	r test (+)	l test	rt est. (−)	g test	r test (+)	l test	r test (−)
IS-1	.05	0.054	0.059	0.041	0.047	0.947	0.985	0.029	0.001	0.047	0.071	0.061	0.284
	.01	0.011	0.007	0.009	0.010	0.822	0.948	0.005	0.001	0.008	0.010	0.015	0.113
IS-2	.05	0.040	0.041	0.058	0.056	0.227	0.564	0.044	0.018	0.056	0.059	0.050	0.123
	.01	0.007	0.009	0.009	0.013	0.067	0.302	0.005	0.002	0.011	0.007	0.012	0.026
IS-3	.05	0.051	0.040	0.053	0.041	0.055	0.049	0.045	0.047	0.049	0.047	0.044	0.059
	.01	0.010	0.014	0.013	0.008	0.010	0.013	0.007	0.013	0.002	0.014	0.008	0.019
SS-1	.05	0.053	0.058	0.053	0.050	0.986	0.996	0.022	0.000	0.063	0.064	0.124	0.582
	.01	0.007	0.011	0.010	0.010	0.960	0.988	0.002	0.000	0.017	0.010	0.041	0.398
SS-2	.05	0.044	0.058	0.058	0.063	0.798	0.909	0.047	0.004	0.034	0.044	0.059	0.316
	.01	0.011	0.011	0.013	0.016	0.676	0.771	0.010	0.000	0.004	0.005	0.014	0.147
SS-3	.05	0.047	0.046	0.057	0.049	0.054	0.128	0.040	0.042	0.044	0.054	0.049	0.071
	.01	0.014	0.007	0.011	0.013	0.014	0.036	0.006	0.010	0.003	0.009	0.006	0.009
IBD-1	.05	0.044	0.050	0.053	0.048	0.962	0.999	0.021	0.000	0.025	0.087	0.438	0.809
	.01	0.008	0.012	0.009	0.010	0.926	0.997	0.003	0.000	0.009	0.023	0.192	0.694
IBD-2	.05	0.052	0.045	0.061	0.038	0.967	0.998	0.023	0.000	0.046	0.076	0.451	0.794
	.01	0.009	0.008	0.011	0.009	0.932	0.997	0.004	0.000	0.009	0.018	0.208	0.672
IBD-3	.05	0.052	0.046	0.053	0.050	0.977	0.999	0.015	0.000	0.050	0.083	0.441	0.824
	.01	0.013	0.009	0.011	0.012	0.939	0.999	0.005	0.000	0.009	0.023	0.225	0.684

* p-values are in italic when non significant and in bold when the fraction of true positive is above 20%
Results show the proportion of significant tests over 1000 replicates, based on 1000 permutations with thresholds .05 and .01

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com