To more directly address the impact of low frequency, non-synonymous genetic variants, we considered only the variants with MAF of <5% (33).
Owing to the common particular interest in discovering loss-of-function variants we considered a heterozygous deletion as a typical scenario for our power estimations (Fig. 2).
To identify possible causal variants we considered 94,783 SNP and 9,662 Indels located within a 10.55 Mb interval on chromosome 10 (25,929,000 bp – 36,479,000 bp).
However, in CAVIAR, in order to detect the ρ causal variants, we consider all possible causal sets which can be very slow depending on the number of variants selected in the ρ causal variant set.
To assign a prior probability for the APE1/Ref-1 -141T/G variant, we considered that the genetic variant has been previously shown to influence expression of the APE1/Ref-1 gene [ 28, 29] and has been previously reported to be associated with lung cancer risk [ 28, 29].
For each variant, we considered a beta-binomial distribution (following Pickrell et al., 2010a) to model the number of reads from the haplotype (denoted as xi+) or the number of reads from the haplotype (denoted as xi−), conditional on the number of total reads (denoted as yi = xi+ + xi−), for each individual i, or x i + | y i ∼ b i n o m i a l (y i, θ ), θ ∼ b e t a.
For example, their more stringent segregation requirements and lack of consideration of functional experimental (e.g. patch-clamp) evidence likely led to their classification of three variants that we considered as "known pathogenic" as "variants of unknown significance", i.e., CACNA1S p.T1354S, SCN5A p.T220I, and SCN5A p.E428K.
In addition, we observed a recurrent sequence variant, which we considered not causative of disease, consisting of a 12 bp sequence in exon 16 c.
For each of the two dissociation variants, we therefore considered two scenarios: In the first, we assume that this control does not exist by setting u' = 1.
To identify possible causal variants for FH4, we considered 43,676 SNP and 4,087 Indels located within a 3.96 Mb interval on chromosome 12 (9,859,000 bp – 13,805,000 bp).
To characterize the variant status, we consider two classes of models: 1) reference model M r, where all variants are explained by sequencing error or contamination; and 2) variant model M v, where variants are explained jointly by sequencing error and the presence of a variant allele at fraction f.
