Provably-good Triangulations for Protein Modeling
Alper Üngör, Department of Computer Science, Duke University
Oct 17, 2003, Wean 7220, 3:30pm

Abstract:

Delaunay triangulations are popular in many fields, including engineering simulations, geographic information systems, computer graphics, visualization, and biological modeling. Triangulations (meshes) are often required to have bounded aspect-ratio and small number of elements. I will present recent results on generating such meshes using incremental point insertion.

In the first part of the talk, I will introduce a new variant of the well-known Delaunay refinement algorithm. In theory, the new algorithm outputs size-optimal and quality-guaranteed meshes, and has better parallel time complexity bounds than the original one. In practice, it runs faster and generates smaller meshes.

The second part of the talk will cover new triangulation algorithms designed to be effective in protein modeling. In particular, challenges in triangulating protein surfaces and computing critical points will be addressed.

Short Biography

Alper Üngör obtained his Ph.D. degree in Computer Science at the University of Illinois at Urbana-Champaign in 2002. He is currently a Visiting Assistant Professor of Computer Science at Duke University. Alper received the David J. Kuck Best Ph.D. Thesis Award, the C.L. Dave and Jane Liu Award, and the Excellence in Teaching Award at the University of Illinois. His research interests include the design and analysis of algorithms, computational geometry, mesh generation, scientific computing, and computational biology.

Alper Üngör
Department of Computer Science, LSRC
Duke University, Durham, NC 27708-0129
Email: ungor@cs.duke.edu
http://www.cs.duke.edu/~ungor