Table 3.

Mutation detection experiment with patient heterozygous for a variant not found in k control individuals (2k chromosomes)

A
			Variant-specific θ for coding regions
	Genome-wide average θ			Base conserved in mice			Base conserved in fugu
k	P-value if L = 10 kb	Max L (kb) for P < 0.05	P-value if L = 10 kb	Max L (kb) for P < 0.05	P-value if L = 10 kb	Max L (kb) for P < 0.05
10	0.44	0.7	0.27	1.4	0.23	1.8
20	0.28	1.3	0.16	2.7	0.13	3.4
50	0.14	3.2	0.07	6.5	0.06	8.2
100	0.08	6.4	0.04	12.9	0.03	16.3
200	0.04	12.8	0.02	25.8	0.02	32.4

A P-values, obtained using equation 2, for 10 kb of sequence examined and maximum length of sequence (kb) that may be examined to keep P-value under 0.05.
B P-values, obtained using simulation, for 10 kb of sequence examined for a population of constant size 10,000 and for a population that has expanded from 10,000 to 6 × 10⁹ over the last 500 generations.
C Minimum number of control individuals (k), obtained using equation 2, that must be examined and found free of the variant of interest to obtain P-value close to 0.05, as a function of length of sequence examined (L). Genome-wide average θ of 8.25 × 10^-4 was used.

On the probability that a novel variant is a disease-causing mutation