Skip to main content

Table 3 Number of SNPs and clumps in simulation study 2 with empirical power > 0

From: r2VIM: A new variable selection method for random forests in genome-wide association studies

       Causal clumps FP clumps
OR n Method Factor Total Causal no no SNPs no no SNPs
1.3 6000 LR   765 9 9 741 2 24
1.3 6000 r2VIM 1 194 8 9 178 13 16
1.3 6000 r2VIM 3 110 8 9 106 3 4
1.3 6000 r2VIM 5 78 8 9 77 1 1
1.1 20000 LR   106 5 5 105 1 1
1.1 20000 r2VIM 1 104 5 8 62 32 42
1.1 20000 r2VIM 3 43 4 7 30 8 13
1.1 20000 r2VIM 5 26 4 7 21 3 5
  1. Shown are results for logistic regression (LR) and r2VIM. Columns denote odds ratio (OR), sample size (n), method, factor for r2VIM, total number of SNPs, number of causal SNPs, number of clumps based on causal SNPs, number of SNPs in clumps based on causal SNPs, number of clumps based on false-positive (FP) SNPs and number of SNPs clumps based on false-positive (FP) SNPs