Skip to main content

Table 5 Power (%) of RAML and SKAT methods to detect association of rare variants under seven scenarios for underlying genetic architecture using data simulated for BRCA2 in 4000 cases and 4000 controls

From: The admixture maximum likelihood test to test for association between rare variants and disease phenotypes

   Threshold for significance
   P < 0.001 P < 0.01 P < 0.05
Scenario* Proportion of variants associated RAML SKAT-O RAML SKAT-O RAML SKAT-O
1 0.05 78.5 14 87.5 22 96 42.5
2 0.05 69.5 8.5 84 18 93 36
3 0.05 67.5 9.5 76.5 21 85 38
4 0.05 61.5 6 72 18.5 83.5 32.5
5 0.05 68.5 8.5 78.5 16 87 35.5
6 0.05 61 12.5 75.5 17.5 82 30.5
7 0.05 65 9 79 21.5 86.5 37.5
1 0.10 75 14.5 86 23 90.5 40.5
2 0.10 61.5 13.5 81.5 22 92.5 39.5
3 0.10 59 9.5 76 20 89 34
4 0.10 58.5 7 72 20.5 84 41
5 0.10 65 9 82 18.5 89.5 39
6 0.10 61 9 79 15.5 86.5 31
7 0.10 66.5 5 83.5 17 91 36
1 0.20 83.5 19.5 95 41 99 63.5
2 0.20 63 9.5 77 22.5 93.5 41
3 0.20 66.5 7 85.5 22.5 94 43.5
4 0.20 53 8.5 72.5 21 87 38.5
5 0.20 70.5 10 86 25.5 93 44
6 0.20 59 11 79 20.5 92 41.5
7 0.20 61 5.5 79.5 18.5 89.5 43.5
  1. Method with greatest power emboldened.
  2. * See text for description of genetic architecture for each scenario.
  3. RAML Rare admixture maximum likelihood, SKAT-O sequence kernel association test, T1 fixed threshold test 1 per cent MAF, T5 fixed threshold test 5 per cent MAF, WST weighted sum test, VTT variable threshold test, EREC estimated regression coefficient test.