We have assembled an interdisciplinary team to address fundamental computational problems in the representation of molecular structures and the simulation of biochemical processes important to life. Among these are ligand-to-protein docking, ab initio structure prediction, and protein folding. We also plan to consider engineering tasks, such as drug and protein design. Through a novel focus on geometric and topological representations, we have an opportunity to enable new scientific breakthroughs and more generally to transform the way we represent, analyze, communicate, and teach those fundamental structures and processes in molecular biology. In the process of doing so we will need to make advances in several areas of information technology that will be scientific accomplishments in themselves as well as being potentially relevant to other natural sciences and engineering disciplines dealing with computer modeling of the physical world.

The BioGeometry project is based on the grant "Computational Geometry for Structural Biology and Bioinformatics." It is supported by the National Science Foundation under grant number CCR-0086013. Any opinions, findings, and conclusions or recommendations expressed on this website are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

The function of all life forms depends on organization in space and time, and the effect of one part of a biological system on another is generally much greater when the two parts are in close proximity in space and/or time. In themselves, these two observations would seem to indicate that geometric methods should be an essential component of any attempt to understand and simulate biological systems. Existing techniques in computational structural biology and bioinformatics, however, rely primarily on sequence information and use statistical and energy-based methods to analyze biological structure and function. They have been developed over three decades and have their roots in methods first applied by computational chemists to much smaller molecular systems. Although there have been significant advancements in the field, a systematic solution of many of the most important biological problems is still elusive, including AB initio protein structure prediction, the protein folding process, and ligand to protein docking. It is widely believed that the geometry of molecules plays a crucial role in these processes, yet geometrical methods are relatively uncommon in computational biology because several unresolved representational and algorithmic issues remain.

We propose to develop new computational techniques and paradigms for representing, storing, searching, simulating, analyzing, and visualizing biological structures. We will rely on geometry, but combine it with statistics and physics. We will aim for methods that have practical, predictive power and validate them by comparison with the best existing techniques. In order to transfer the technology in an effective way and have real impact on research in biology, the proposed project will create software whose aim is to help structural biologists with their research and integrate with their current tools.

Ideas from a wide range of areas of computer science and mathematics, including algorithms, geometry, topology, graphics, robotics, and databases will be needed to accomplish our goals. Some of the problem areas to be addressed represent great challenges for computer science itself. These include building and querying large libraries of three-dimensional and possibly flexible shapes, exploring hierarchical representations of deformable geometry, integrating geometry and physics in modeling, and properly sampling systems with many degrees of freedom.

Although we focus on computational structural biology and bioinformatics applications, the research carried out under the proposed project will have a wider impact and will advance knowledge in several areas of computer science, including geometric modeling, shape analysis, 3D geometry databases, physical simulation, robotics, and visualization.

The project will also foster integration of research and education in the proposed domains. We have put together a team of researchers with strong credentials who have already made significant contributions to these and related areas. To ensure that we address the real world problems in computational structural biology and bioinformatics and that we have a fruitful collaboration, biologists, chemists and physicists will participate and be involved in the research from the beginning.

This proposal has aspects of both the Advanced Computational Science and the Information Management parts of the ITR solicitation, as it addresses the need for new geometric representations and algorithms in structural biology.

Article in Duke Daily Dialogue
On March 2, 2001, the Duke Daily Dialogue featured the BioGeometry project in an article titled "Researchers 'Shaping' Future of Proteomics." The article is not currently available on the Dialogue site, but a .doc file of the text is available. Click here to see it.