Microsatellite mutations in the germline: implications for evolutionary inference

Microsatellite DNA sequences mutate at rates several orders of magnitude hi gher than that of the bulk of DNA. Such high rates mean that spontaneous mu tations that form new-length variants can realistically be seen in pedigree analysis. Data on observed mutation events from various organisms are now accumulating, allowing inferences on DNA sequence evolution to be made thro ugh an unusually direct approach. Here I discuss and integrate microsatelli te mutation data in an evolutionary context. A striking feature of the muta tion process is that it seems highly heterogeneous, with distinct differenc es between species, repeat types, loci and alleles. Age and sex also affect the mutation rate. Within genomes at equilibrium, the microsatellite-lengt h distribution is a delicate balance between biased mutation processes and point mutations acting towards the decay of repetitive DNA. Indeed, simple repeats do not evolve simply.