CRePE Project 4:
Computational and
Mathematical Analysis
of Biomedical Data
Our ultimate
objective is to predict, detect, and
explain microbial adaptation, viewed as
an evolutionary phenomenon, since
evolution is a major factor in the
emergence and spread of disease as well
as in the large scale effectiveness of
many treatments. This project will move
towards this objective by meeting the
following specific aims:
-
Develop new algorithms for aligning
multiple sequences and inferring
phylogenies;
-
Evaluate the accuracy of new and current
algorithms for inferring phylogenies,
detecting recombinations, and aligning
sequences;
-
Apply mathematical models of spatial organismal interaction and sequence loci
interactions to evolutionary data from
controlled experiments.
Our new algorithms
include: an iterative approach to
discovering subtle similarities in
subsequences with which to guide the
full alignment; and a genetic algorithm
to guide a progressive dynamic
programming alignment. The phylogenetic
inferencing algorithms use genetic
algorithms to search the vast space of
possible trees more efficiently. We
evaluate representative current
algorithms, as well as our own, using
data from experimental evolution in
other projects, and powerful statistical
techniques for modeling recombination
events. For the first time, we apply
mathematical tools to spatial
constraints in viral evolution, and we
use mathematical tools from the theory
of evolutionary computation to
investigate the emergence of correlated
motifs in protein evolution and in
abstract evolutionary processes.
Contact:
Dr. James Foster (Computer
Science)
|
