Erratum: Broman KW. 2001. Estimation of Allele Frequencies With Data on Sibships. Genet Epidemiol 20:307–15.
✍ Scribed by Karl W. Broman
- Publisher
- John Wiley and Sons
- Year
- 2002
- Tongue
- English
- Weight
- 55 KB
- Volume
- 23
- Category
- Article
- ISSN
- 0741-0395
No coin nor oath required. For personal study only.
✦ Synopsis
In the April 2001 issue of Genetic Epidemiology, in the article ''Estimation of Allele Frequencies With Data on Sibships,'' by , there is an error on page 310, in the second paragraph under ''Method 3: Accounting for Relationships.'' The stated probabilities that an allele in the second sibling is not identical by descent (IBD) with one of the first sibling's alleles, written as p l /(1+p l ), are incorrect; we had missed two important cases. Let (g 11 , g 12 ) denote the two alleles of the genotype of the first sibling, (g 21 , g 22 ) denote the two alleles of the genotype of the second sibling, and g ¼ (g 11 , g 12 , g 21 , g 22 ). Further, let pg denote the genotypes for the two parents, and A denote the event ''g 21 is not IBD with g 11 or g 12 .'' We seek Pr(A|g), which we calculate by conditioning on the parents' genotypes, as follows: PrðAjgÞ ¼ P pg PrðpgÞ PrðgjpgÞ PrðAjg; pgÞ P pg PrðpgÞ PrðgjpgÞ
The correct probabilities appear in Table I.
As a result of this error, the numbers in Table I of Broman (2001) were slightly wrong. We have rerun our computer simulations, after modifying our algorithm using the corrected probabilities, to obtain the true maximum likelihood estimates (MLEs). The corrected version of the table appears in Table II.