Land Resources & Environmental Sciences Seminar Series
- Monday, March 2, 2015 from 1:10pm to 2:00pm
- Leon Johnson Hall - view map
Jason M. Wood, LRES Graduate Student and IGERT Intern, will present "Ecotype Simulation: Demarcation of bacterial species from DNA sequences using ecological theory" as part of the LRES Seminar Series in 346 Leon Johnson Hall. Abstract: The identification of closely related, ecologically distinct populations within a microbial community is complicated by the lack of phenotypic differences between populations. Historically, the field of microbiology has advocated the use of molecular cutoffs (similarity in DNA sequence) of highly conserved genes to demarcate species. However, if the same cutoffs are used within the Hominidae lineage, all of the great apes, including humans, would be grouped into a single species! This has led to a theory-based approach to the identification of microbial species based on the Stable-Ecotype Model of species and speciation that model DNA sequence diversity resulting from net ecotype (species) formation (Darwin's divergence of character) and periodic selection (Darwin's struggle for existence). Ecotype Simulation is a cross-platform program that utilizes the Stable-Ecotype Model in a Monte-Carlo style simulation to determine the most likely rate and timing of ecotype formation and periodic selection events to explain the observed molecular phylogenies. Ecotypes predicted by Ecotype Simulation have been shown to be ecologically distinct, with Synechococcus ecotypes distributing themselves along fine-scale temperature and light gradients found in a microbial mat growing along the effluent channel of an alkaline, siliceous hot spring in Yellowstone National Park. In collaboration with colleagues who developed the original version of Ecotype Simulation, I have helped develop a significantly faster version of the program to handle the large increase in the amount of DNA sequence provided by modern sequencing techniques, extending the number of sequences that can reasonably be analyzed from approximately two hundred to potentially one million. I utilize Canonical Correspondence Analysis to visualize the distribution of predicted ecotypes within an ordination space of the niche-defining environmental parameters, and to compare the difference in ecotype demarcation between the two versions of Ecotype Simulation.