BCB at ISU- Speaker Information

Ling Guo

Bioinformatics & Computational Biology Student Seminar Series

Greedy Closure Evolutionary Algorithms

Ling Guo
Major professors: Dan Ashlock and Pat Schnable
Iowa State University Friday, March 29, 2002
1:10 p.m.
1420 Molecular Biology Building

Abstract
Greedy algorithms are familiar to most software developers. A few, like the algorithm for finding a minimal-weight spanning tree, can be proven to yield optimal results. Problems such as graph coloring or the traveling salesman problem admit a plethora of greedy algorithms which yield sub-optimal results. The control of greedy algorithms seems a natural target for evolutionary computation but relatively few papers have appeared. There are several possible approaches. The approach explored here consists of making small modifications in the order of presentation of potential parts of a growing structure as a means of deflecting the greedy algorithm's behavior. The role of evolutionary computation is to locate the modifications of the order of presentation. We demonstrate this technique on the location of embeddable DNA markers for genetic libraries. The technique has the potential for broad application.

Huaijun Zhou

Bioinformatics & Computational Biology Student Seminar Series

Genetic control of immune response in livestock

Huaijun Zhou
Major Professor: Dr. Xun Gu
Iowa State University Friday, March 29, 2002
1:10 p.m.
1420 Molecular Biology Building

Abstract
The immune system plays a very important role in providing protection from a wide range of pathogens. There are two distinct cellular systems: the innate immune system and the acquired immune system. The innate immune system acts effectively without previous exposure to pathogens. The Toll-like receptor gene family plays important role in signal transduction pathway in the innate immune response. Current approaches are not sufficient to investigate functional genomics with gene family. Functional divergence analysis with statistical model is cost-effective approach to conduct this kind of study. The amino acid sequences of all Toll-like receptor gene family were collected, and multiple aligned. The phylogenetic tree was inferred, and critical amino acid residues were predicted by functional divergence analyses. The results indicate that the site-specific profile based on posterior analysis is useful not only for understanding the functional basis of protein family evolution but also a cost-effective approach in functional genomics.