Robust methods for population stratification in genome wide association studies

BMC Bioinformatics

Table 2 Empirical false positive rate and true positive rate results for simulation I (Discrete Populations without Outliers)

Case control	Testing SNP	Trend	GC	PCA	MDS	RPCA-	RPCA-
Difference	Types					RHM	PP
S1	Random SNPs	2.67	0.91	0.97	0.97	0.99	0.97
(2 populations, moderate)	Differentiated SNPs	99.85	98.86	1.30	0.90	0.88	0.89
(2 populations, moderate)	Causal SNPs	48.99	34.13	47.37	47.29	46.92	47.33
S2	Random SNPs	16.56	0.89	1.11	0.92	0.93	0.92
(2 populations, more extreme)	Differentiated SNPs	100.00	100.00	13.60	1.00	1.01	0.99
(2 populations, more extreme)	Causal SNPs	49.91	10.91	33.89	31.76	31.63	31.77
S3	Random SNPs	3.14	0.97	0.94	0.93	0.95	0.92
(3 populations, moderate)	Differentiated SNPs	99.99	99.98	2.24	1.00	1.01	1.00
(3 populations, moderate)	Causal SNPs	48.18	31.76	45.16	45.08	44.60	45.09
S4	Random SNPs	21.76	0.94	1.45	1.05	1.05	1.06
(3 populations, more extreme)	Differentiated SNPs	100.00	100.00	41.78	0.96	0.95	0.96
(3 populations, more extreme)	Causal SNPs	50.79	8.42	23.51	19.34	19.13	19.34

^aFor random SNPs and differentiated SNPs, the values in the table represent the empirical false positive rates; for causal SNPs, the values in the table represent the empirical true positive rates. The nominal false positive rate is 0.01. Note that the numbers in the table refer to percentages.

ISSN: 1471-2105