Comparison of collapsing methods for the statistical analysis of rare variants

BMC Proceedings

Table 1 Minimal power of five collapsing approaches

SNP type	MAF	WS	CMC	CAST	RVT1	RVT2
All	0.01	0.015	–	0.010	0.010	0.010
	0.02	0.025	0.030	0.030	0.040	0.030
	0.03	0.070	0.050	0.050	0.060	0.055
	0.05	0.035	0.050	0.050	0.060	0.050
	0.07	0.120	0.115	0.115	0.125	0.115
	0.10	0.135	0.075	0.075	0.115	0.070
Synonymous	0.01	0	–	0	0	0
	0.02	0.005	0	0	0	0
	0.03	0.005	0	0	0	0
	0.05	0	0	0	0	0
	0.07	0.005	0.005	0.005	0	0.005
	0.10	0	0	0	0	0
Nonsynonymous	0.01	0.010	–	0.010	0	0.010
	0.02	0.050	0.045	0.045	0.025	0.045
	0.03	0.135	0.120	0.120	0.090	0.120
	0.05	0.130	0.120	0.120	0.090	0.120
	0.07	0.285	0.170	0.170	0.285	0.170
	0.10	0.235	0.170	0.170	0.285	0.170

Minimal power of all five collapsing approaches for different SNP types and MAF thresholds to control a family-wise error rate ≤ 0.05. WS, weighted-sum method; CMC, combined multivariate and collapsing method; CAST, cohort allelic sum test; RVT1, rare variant test with proportion coding; RVT2, rare variant test with indicator coding.

ISSN: 1753-6561