Marker Density and Estimates of Chromosome Rearrangement

Abstract

The number of chromosomal breakpoints that have occurred since divergence of two species from their common ancestor may be estimated by comparing the map positions of homologous marker loci. The random breakage model has served as the basis for several such estimates. The model treats detection probabilities for translocation and inversion as equal, and assumes that chromosome boundaries and breakpoints are distributed randomly along a single interval of the genome. Simulations and numerical analyses were used to test the validity of these assumptions. The estimated number of breakpoints is close to expected under the model when marker density is high relative to the amount of chromosomal rearrangement, but it is biased downward when marker density is low, especially when chromosomal evolution occurs by inversion. The bias arises because rearranged segments may contain markers yet, nevertheless, be undetected. Variances of the breakpoint estimates decrease rapidly as markers are added to the comparative maps. Variances obtained when chromosome length variation is constrained are substantially lower than those obtained when chromosome size is assumed to follow a uniform random distribution.