Mutation is the key player in evolutionary change, both in micro and macro scenarios. Defined as any hereditary change in the transmitted genome from progenitor to offspring, it embraces a heterogeneous set of phenomena, in terms of molecular bases and consequences. Here we limit our analyses to those corresponding to site-specific changes in nucleic acid sequence composition, without alteration in the genome size, i.e., substitutions. These analyses will be performed in the framework of the classical Mendelian diploid mode of transmission, without sex-linkage, under a simple Hardy-Weinberg model. Under these conditions, we show that: (i) per site mutation rates are evolutionarily uninformative, biallelically defined parameters being required to predict the genetic behavior of the population and are also essential to medical and forensic applications; (ii) at polymorphic sites (heterozygosity > 0), estimates of mutation rates through the simple counting of Mendelian incompatibilities in pedigrees are always underestimations, depending heavily on the allelic frequencies, a fact that may lead to erroneous evolutionary inferences, such as that (iii) apparently equal forward and reverse mutation rates, as estimated from Mendelian incompatibilities, do not equate to stable equilibrium. An approach allowing to correct the underestimation, by adopting a sequential strategy of allele specific estimation is presented.

mutation rate estimation, substitution, population genetics, evolution