F1000 evaluations of Markova and Petrov

Obtaining genomic data from large numbers of species has now become a feasible reality. In comparative genomics, these data serve to understand the way in which DNA sequences are affected by different evolutionary processes (e.g. selection, genetic drift and mutation bias). One important question that can be addressed is what fraction of genes and nucleotide sites have been subject to positive selection. However, a critical step to determine this is to obtain reliable alignments of orthologous sequences. While manual editing of sequence alignments after running programs was typically standard before genome-wide datasets, this is no longer feasible given the scale of genomic datasets, and we are increasingly reliant on the accuracy of automated aligners.