Computer Science 597
Advanced Topics in Computer Science
Computational Biology
Tandy Warnow
Fall 1997
Seminar on Computational Biology
Topic: evolutionary tree reconstruction methodology and
algorithms in Biology and Linguistics.
Prerequisites: graduate course in discrete algorithms
and interest in experimental algorithms and scientific
issues.
No biological or linguistic knowledge will be presumed.
Course will be held on Tuesdays from 10-12 in 301 Computer
Science. This is a change from the original schedule!
September 15: introduction.
September 23.
Distance-based optimization problems
and algorithms. Additive and ultrametric trees.
Buneman four-point theorem.
Combinatorics of trees and character encodings of tree topologies.
Quartet-based tree reconstruction algorithms.
Topology invariant neighborhoods of additive trees.
September 30.
Farach-Kannan-Warnow optimal ultrametric tree reconstruction and
related algorithms.
Agarwala et al. 3-approximation algorithm for
the -nearest tree problem.
October 7.
Stochastic models of iid site evolution, and performance of
distance-based methods for trees under such models.
Statistical performance issues.
October 14.
Character based tree reconstruction methods.
Parsimony, compatibility, and maximum likelihood estimation.
NP-hardness and approximation algorithms for parsimony
and compatibility.
Statistical properties of character-based methods.
October 21.
Perfect Phylogeny problem and algorithms.
Triangulating colored graphs formulation.
November 4.
Combinatorial algorithms for
fixed parameter perfect phylogeny.
November 11.
Graph theoretic algorithms for
fixed parameter perfect phylogeny.
November 18.
Tree reconstruction in Historical Linguistics, and
application of perfect phylogeny algorithms.
November 25.
Consensus tree problems and algorithms.
December 2.
Experimental performance analysis in biological systematics.
Open questions and pressing debates.
The role of experimental studies (why analytical results
are not enough) and methodological problems.
December 8.
Other topics, as interest and time permits.
Possible topics:
Multiple sequence alignment, short quartet methods, ``real
data", open problems in historical linguistics.
Instructor: Tandy Warnow, tandy@cs.princeton.edu.
Office 206 Computer Science. Phone 258-2211.
Office hours to be announced.