TEXTAL is an automated system for protein model-building in X-ray crystallography that is based on AI and pattern recognition methods. X-ray crystallography is the most widely used method for determining the three-dimensional structures of proteins and other macromolecules. Interpreting the 3D image of the electron density surrounding a protein has traditionally been done manually and is very time-consuming and error-prone, requiring human judgement and background knowledge to compensate for limited resolution and noise
from various sources. TEXTAL is designed to automate this process by predicting atomic coordinates using feature-based analysis of patterns in electron density maps. TEXTAL uses a variety of AI and pattern recognition techniques to try to mimic the intuitive decision-making processes of experts, for building both backbone and side-chains, in solving protein structures. The system has proven successful in determining protein structures over a range of intermediate resolutions. In this talk, we will give an overview of TEXTAL and describe some of the algorithms it uses. TEXTAL has been developed over the course of 8 years through a collaboration between researchers
in the Computer Science Dept. and the Dept. of Biochemistry and Biophysics at Texas A&M, and is now used by structural biologists throughout the world.
Dr. Ioerger is an Associate Professor in the Department of Computer Science at Texas A&M. He received his BS degree in Molecular and Cell Biology from Penn State in 1989, and an MS and PhD in Computer Science from the University of Illinois in 1992 and 1996, respectively. His research interests are in Artificial Intelligence and Bioinformatics.