Determination of sample size for a multi-class classifier based on single-nucleotide polymorphisms: a volume under the surface approach

Table 2 Sample size determination: here, D = 3 and 4 , and n is the sample size for each class satisfying: $\hat{VUS (\infty)} - \hat{VUS (n)} < γ (= 0.01)$

Here, ${\vec{θ}}_{1} = {(θ_{1, 1}, \dots, θ_{1, m})}^{'}$ , let θ_1,j∼U(0.4,0.49),j=1,…,m; for a specified scalar value h, let ${\vec{h}}_{1}, {\vec{h}}_{2}, {\vec{h}}_{3}$ be such that their components h_i,j∼U(h−0.002,h+0.002), j=1,…,m; and let ${\vec{θ}}_{i + 1} = {\vec{θ}}_{i} - {\vec{h}}_{i}, i = 1, 2, 3$ ; m is the number of independent SNPs, α=0.01 is the significant level for Wald tests; and ρ=1 is the percentage of the significant SNPs.

ISSN: 1471-2105