Current Position: Associate Professor, Computer Science
Research Interests: The design of efficient, scalable, and parallel algorithms for various data mining techniques. Professor Zaki is specially interested in developing novel data mining techniques for bioinformatics.
Contact Map Mining for Protein Structure Prediction: Given a protein amino acid sequence (linear structure), determining its three dimensional folded shape (tertiary structure) is referred to as the Structure Prediction Problem, one of the grand challenges in Bioinformatics. Instead of traditional simulation studies, I am learning to predict the structure from known protein structures (e.g., Protein Data Bank), using Protein Contact Maps (two dimensional representations of the three dimensional structure of proteins). My work is targeting two main problems: 1) Given a database of protein sequences and their 3D structure in the form of contact maps, build a model to predict if pairs of amino acids are likely to be in contact or not. 2) Discover common (non-local) contact patterns or ``features'' that characterize physical ``protein-like'' contact maps.
Education Professor Zaki received his Ph.D. degree in computer science from the University of Rochester in 1998.
Teaching: Data Mining, Computational Biology & Bioinformatics, Database Systems
Current Students: Vineet Chaoji, Mohammed Al Hasan, Saeed Salem, Abhinab Ray