I want to bring your attention to a somewhat dense and possibly inconclusive (but important) paper accompanied by a very informative overview in PLoS Biology, concerning mutations in the human genome.
Mutation rates and patterns of mutation are important for a number of reasons. For one thing, the genome itself is a data set that is both broad and deep. There is a lot of information in a given individual genome (a haploid set of genes from a person, for instance) but there is a wide range of variation in that information. So, inferences or assertions regarding the nature and distribution of genes or their variants cannot really refer to a single version of the genome, but must also take into account the variation in DNA sequences.
A very obvious area where variation is important is in reconstructing phylogenies. “Family trees” of populations or species can be reconstructed by estimating the genetic difference between pairs of samples, and from this, estimating the amount of time that has passed between a Last Common Ancestor and each of two later populations. These dyads (or triads, depending on how you count them) can then be pieced together to get a phylogeny … a graph representing the historical divergence of populations or species … that tells us a particular version of history. Obviously, the rate of mutation must be known or assumed to make this work. Variation in mutation across the genome, or across a population, or across the structure of the family tree itself will cause incorrect inferences.
The research paper is “Cryptic Variation in the Human Mutation Rate” by Hodgkinson et al. Here’s the key finding:
Continue reading Mutation Patterns in the Human Genome are More Variable Than Expected →