1998 Volume 73 Issue 6 Pages 337-343
Distorted segregation has been repeatedly observed in molecular marker linkage mapping of various plants and animals where distant crosses were made. It may be caused by a semi-lethal factor acting in the filial generations. Although general methods have been developed to find semi-lethal factors, so far no method is available to estimate semi-lethal factors in the organisms, for example Bombyx mori and Drosophila melanogaster, in which no crossing-over occurs in meiosis in either of the sexes. In this report, a method is presented for estimating the recombination values between a semi-lethal factor locus and neighboring dominant molecular markers, e.g. RAPDs and AFLPs, and the relative viability of gametes or zygotes affected by the semi-lethal factor in an F2 population for such species by using the maximum likelihood method associated with the expectation maximization (EM) algorithm. Five selection models of gamete or zygote were considered, and the most likely one was determined by goodness of fit of the observed counts of phenotypes to the expected ones under the models. The method was applied to segregation data of RAPD markers of an F2 population with 101 individuals developed from the `C108' × `p50' cross in silkworm, B. mori. The presence of two semi-lethal factor loci (L 1 and L2) located on the linkage groups 22a and 27b causing partial selection was suggested. The semi-lethal factors L1 and L2 were theoretically predicted to affect viability of male gametes and zygotes, respectively.