/sites/default/files/styles/banner_image/public/default_images/inside-page-banner_2_1.jpg?itok=Er8q0C-3
Associate 2014-15

James Whitfield

Entomology

Whitfield imageTesting of new phylogenetic network methods with appropriate empirical biological data sets

Since the days of Darwin, the prevailing metaphor for the evolution of life on Earth has been that of a phylogenic tree, with the trunk representing the earliest life on Earth, branches representing the evolutionary lineages of life forms, and the world’s present-day species depicted on the tips of branches.

Not all of evolution, however, produces tree-like historical patterns. It is now being realized how complex genomes really are, both in their composition and in their evolutionary histories. Genetic recombination, gene conversion between paralogous gene copies, operon formation, lineage sorting among alleles – these and other genetic phenomena can produce conflicting patterns that cannot be summarized effectively with a single evolutionary tree. A promising alternative is using phylogenic networks to help reconstruct relationships among organisms and interpret their genomic data.

During his Center appointment, Professor Whitfield will serve as biological problem collector in an international collaboration with mathematicians, computer scientists, and other biologists to develop practical network methods that answer the questions biologists are asking. He will supply real biological datasets to test several of the mathematicians’ network approaches. For example, Eubacteria and Archaea display an unusually high level of gene-sharing, with complex genomic relationships. Network visualization tools for these relationships are already under development, but how will the methods scale up to larger problems as new sequenced genomes accumulate? Only tests with real data will tell.

In addition to validating or invalidating the network methods being developed, the project will result in a greatly expanded research base of test datasets that will benefit network research for years to come. It is also likely that the collaboration will lead to new network methods not yet conceived.