The equilibrium distribution for a generalized Sankoff-Ferretti model accurately predicts chromosome size distributions in a wide variety of species (Q2748430)

This paper is concerned with the distribution of chromosome sizes in various species. In the evolution of genomes, there are three types of change: NEWLINENEWLINENEWLINEa) Fission and fusion: chromosomes split into two, or two chromosomes combine into one; b) Translocation: part of the ends of two chromosomes break off and re-attach to the other chromosome; c) Inversion: a chromosome breaks in two places, and on re-assembling, the central piece has changed its orientation. NEWLINENEWLINENEWLINE\textit{D. Sankoff} and \textit{V. Ferretti} [Genome Res. 6, 1-9 (1996)] considered a stochastic model for the evolution of the size of \(k\) chromosomes due to random reciprocal translocations, but ignoring the existence of centromeres. They found that when their model was fitted to data, the short chromosomes were too short and the long ones too long. NEWLINENEWLINENEWLINEThe present work considers a generalization of their model with explicit centromeres, and the introduction of fitness functions based on recombination probabilities and on the length of the longest chromosome arm. A simple formula is derived for the stationary probability of the Markov chain model, which fits the data on chromosome lengths much better than the model of Sankoff-Ferretti, in many but not all of the species considered. Several illustrative graphs are provided to show the fit to data from humans, rats, pigs, yeast, mice, sheep, wheat and rice.