Geometric Computations in Molecular Biology

Abstract

This seminar explores how to represent and manipulate molecular structures (primarily proteins and nucleotides) in software. Students will learn how existing programs tackle problems in structure analysis, energy evaluation, molecular docking, and protein design, and explore new algorithmic solutions.

Objective

The goal of the seminar is to enable cross-disciplinary communication: that each student will be able to learn enough of the motivation, terminology, problems, and existing applications of geometric computation in structural molecular biology that they will be able to contribute to the solution of problems at the interface. We expect to encounter many topics for possible further research projects.

Content

Although the disciplines of biology and computer science have in the past been poles apart, and students of one discipline have traditionally avoided the other, some of the most exciting developments in science are at their interface. One example is the problem of protein folding - how a sequence of amino acids that is coded for by a gene folds into its three-dimensional structure to perform the processes of life. This is arguably the most intriguing puzzle in science today. The best software solutions known for this problem are built on models that are surprisingly geometric: atoms are often represented as hard spheres and molecules as kinematic chains with rotating joints. In this seminar course we will explore these models, their use in existing software such as the Rosetta suite from the Baker lab at the University of Washington, and possible extensions to new algorithms or applications.

Resources