The genotype-phenotype map is an object of central importance in developmental and evolutionary biology. It characterizes the relationship between genotype and phenotype using networks, in which nodes correspond to genotypes labeled by phenotype, and edges connect genotypes that differ by a single small mutation.
By understanding the architecture of genotype-phenotype maps, we can understand the kinds of phenotypic changes mutation can cause. Our work in this area focuses on computational models of gene regulatory circuits, public data describing interactions between regulatory proteins and nucleotide sequences, experiments with synthetic gene regulatory circuits, and the genetic code.
Sewall Wright’s metaphor of the adaptive landscape has framed evolutionary thought for nearly a century. An adaptive landscape is itself a genotype-phenotype map, where the phenotype is fitness or a related quantitative trait, which defines the “elevation” of each coordinate in genotype space. Evolution can be thought of as a hill-climbing process in this landscape, where populations ascend peaks as a consequence of mutation, recombination, and natural selection.
Recent technological advances have elevated the metaphor of the adaptive landscape from a useful abstraction to a measurable property of biological systems. Our work in this area focuses on experimental data from chip-based assays of transcription factor-DNA interactions, as well as from massively parallel sequence-to-function assays for proteins.
Mutation is a biased stochastic process, with some kinds of genetic changes occurring more frequently than others. It is becoming increasingly appreciated that such biases can influence adaptive evolution.
Our work in this area intersects with our work on adaptive landscapes, using modeling to understand how mutation bias might steer an evolving population toward one adaptive peak or another. It also uses experimental data of genetic changes associated with adaptation, derived from laboratory evolution experiments and clinical isolates.
For example, we have shown that mutation bias influences which genetic changes drive the evolution of antibiotic resistance in the global human pathogen Mycobacterium tuberculosis.