SNP calling by sequencing pooled samples

BMC Bioinformatics

Table 1 Priors for f = 0, f = 1, and 0 < f < 1

	Informative	Flat
unfolded	$\{\begin{matrix} 1 - θβ - D f = 0 \\ D f = 1 \\ \frac{θ}{100 f} 0 < f < 1 \end{matrix}$	$\{\begin{matrix} 1 - θ - D f = 0 \\ D f = 1 \\ \frac{θ}{99} 0 < f < 1 \end{matrix}$
folded	$\{\begin{matrix} \frac{(1 - θβ)}{2} f = 0 \\ \frac{(1 - θβ)}{2} f = 1 \\ \frac{θ}{200 f (1 - f)} 0 < f < 1 \end{matrix}$	$\{\begin{matrix} \frac{1 - θ}{2} f = 0 \\ \frac{1 - θ}{2} f = 1 \\ \frac{θ}{99} 0 < f < 1 \end{matrix}$

We discretize the interval [0,1] with N_d= 100 breakpoints. The numbers 99, 100 and 200 appearing in the formulæ in the table are normalization factors (respectively N_d− 1, N_dand 2N_d). β is a normalization constant for the divergent function 1/f, $β = \sum_{i = 1}^{N_{d}} \frac{1}{i} = ln (N_{d}) + γ$ where γ = 0.57721… is the Euler-Mascheroni constant.

ISSN: 1471-2105