Comparative and Evolutionary Genomics

Motivation: Genomes contain the information for constructing organisms. In some sense, this information is put in by selection, and it is degraded by mutation and genetic drift. It is natural to pose some basic questions of principle. What is the maximum amount of information that can be maintained at mutation-selection equilibrium? Can organisms in principle become arbitrarily complex, or, for a given mutation rate and intensity of selection, is there a limit to the amount of information that can even in principle be maintained in the genome? What method of encoding information in the genome allows the greatest amount of information to be maintained for the least selective cost? Indeed, how can the “information from selection” even in principle be defined? As far as we are aware, these questions have not previously been satisfactorily posed nor answered. Results: We make these questions precise and answer them by modelling selective breeding as a communication channel. We show how the information-theoretic capacity of this channel provides a measure of the amount of information that can be put into or maintained in the genome by selection. The channel capacity is computed for some simple genetic models. A striking result is that for a large population in mutation-selection equilibrium, the amount of information that can be maintained using a diffuse encoding analogous to an error-correcting code is vastly greater than the amount of information that can be maintained if information is encoded as exact sequences of nucleotides.