A A A


Speaker Title Date Time & Place
Ariel Cintron-Arias, Department of Mathematics
ETSU
Parameter Estimation from Epidemic Data September 18 3:30 Gilbreath Hall, Room 304
James Livingston, Interim Chair, Biostatistics and Epidemiology, ETSU An Introduction to the Epidemiology of Infectious Disease October 2 3:30 Gilbreath Hall, Room 304
Predictive Modeling Class, Special Topics in Mathematics Predictive Models for an Application in Biomedical Engineering October 23 3:30 304 Gilbreath Hall
Rhydon Jackson, Graduate Student in Computer Science Predicting Flavonoid UGT Regio-selectivity with Graphical Residue Models and Machine Learning November 13 3:30 304 Gilbreath
TBA TBA November 30 TBA

Upcoming Seminar:

Predicting Flavonoid UGT Regio-selectivity with Graphical Residue Models and Machine Learning

Presented by:
Rhydon Jackson, graduate student in Computer Science

Time: 3:30 PM, Friday, November 13, 2009

Location: Rm. 304, Gilbreath Hall

Abstract: Machine learning is applied to a challenging and biologically significant protein classification problem: the prediction of flavonoid UGT receptor regio-selectivity from primary sequence. Novel indices characterizing graphical models of protein residues are introduced. The indices are compared with existing amino acid indices and found to cluster residues appropriately. A variety of feature selection schemes employing the indices are then investigated by examining their cross validation performance when analyzed using nearest neighbor, support vector machine, and Bayesian neural network classifiers. Improvements over nearest neighbor classifications relying on standard alignment similarity scores are reported.