Next generation sequencing technologies have provided us with a wealth of information on genetic variation, but predicting the functional significance of this variation is a difficult task. This thesis summarises the development of a profile hidden Markov model based approach we call deltabitscore (DBS). The approach identifies orthologous proteins that have diverged at the amino acid sequence level in a way that is likely to impact biological function. I present the benchmarking of this approach using several widely used datasets and its application to various biological questions. I then outline the extension of this method from a pairwise comparative method to one that can be scaled for the comparison of hundreds or thousands of bacterial genomes. I demonstrate the utility of the method for identifying associations between genetic variation and phenotypes of interest, and discuss methodological considerations and extensions that must be made in order for this approach to function effectively on a large scale.