Imran Haque

ihaque@cs.SCHOOLNAME.edu

S296 James H. Clark Center
318 Campus Dr W
Stanford, CA 94305-5447

Who am I?

I'm a second-year Ph.D. student in the Computer Science department at Stanford. I am a member of the Folding@Home research group, and am primarily advised by Prof. Vijay Pande. I am co-advised in the CS department by Professor Daphne Koller.

Before that, I graduated from the University of California, Berkeley, with a degree in Electrical Engineering and Computer Science (Go Bears!). I did undergraduate research with Professors Kathy Yelick, Bora Nikolic, and John Wawrzynek. I was also a member and officer for several semesters at the Berkeley Mu Chapter of Eta Kappa Nu.

Even further back, I graduated from Bellarmine College Preparatory in San Jose (Go Bells!). I doubt any high school students will care to read this page, but if you do, I strongly encourage you to do (as I did), speech and debate. Without a doubt, the skills I gained there have been extremely useful to me.

What do I do, and with whom have I done it?

Current Projects

Computational Drug Design - Computational techniques to design small-molecule agents for pharmaceuticals and chemical biology.

Current state-of-the-art experimental techniques for drug development typically include a "high-throughput screening" (HTS) step in which (hundreds of) thousands of compounds are simultaneously tested for activity against a desired target. These experimental screens are labor-intensive, expensive, and time-consuming. I am interested in accurate computational approaches to improve this procedure. The availability of structural information for relevant drug targets, combined with data about the interaction networks present within biological organisms, may make it possible to specifically design chemical agents with higher potency and lower toxicity. These methods are applicable not only to the design of pharmaceuticals, but also to the design of agents to interact with specific cellular systems for research work in chemical biology. I am particularly interested in combinations of the following techniques:

Molecular docking:Docking techniques (also known as virtual high-throughput screening, or vHTS) trade accuracy for speed, with the goal of being an in silico alternative to wet-bench based HTS methods. I am interested in improving the accuracy of vHTS techniques, primarily through improved scoring techniques.
Free-energy calculations:The Gibbs free energy, as applied to molecular design, is a physical parameter that determines the interaction affinity between a chemical agent and its target (colloquially speaking, how strongly the two "stick to" one another), which is a critical determinant of the potency of a particular chemical. I am interested in improving the accuracy and performance of free-energy techniques to make them more applicable to drug design.

Data Visualization and Democratization:

There's an awful lot of information out there that could be of use to the public at large, but for various reasons is largely inaccessible. Obstacles like poor organization and expensive proprietary software should not be restrictions on access to public data by the public.

I wrote an online interface to the data of the US Census using Google Earth because I thought that the existing free tools were simply not powerful enough. Furthermore, the software I developed to run that page is free and open source, unlike the GIS tools used by most professionals. This allows you to map geographic data sets (the Census, geographic sales figures, demographics, etc.) in a very powerful interface for free.
I have also written a simple web application to transform geotagged images in a variety of formats (especially GeoTIFF) into Google Earth KML and KMZ files. Google Earth Pro offers this feature, for a fee, but it's a very simple process. gCensus-GT lets you do it for free.

Past Work

Computational predictive structural biology - computational techniques to solve problems of biological structure.

Examples include the protein folding problem and its relatives, in which one wants to find the 3-dimensional structure of a biological macromolecule given its monomer sequence. In addition to the goal of finding the structure itself, there is the important question of folding mechanics, in which we would like to understand how a protein reaches its final shape, not merely what its final shape is.

Bioengineering 131 and 143, UC Berkeley (Fall 2004/Spring 2005)
Summer 2005 research internship at the IBM Almaden Research Center with Jed Pitera and Bill Swope. Research on folding mechanics, by examining of the folding of a model system.

"Absence of reptation in the high-temperature folding of the trpzip2 beta-hairpin peptide". JW Pitera, I Haque, WC Swope. Journal of Chemical Physics 124, 141102 (2006).

Computational design of biological structure

This can be seen as the inverse of the above problem. In this case, we would like to (for example) design protein sequences which fold into a desired shape. A solution here would have important implications in the design of protein drugs, in the re-engineering of biological function (i.e., synthetic biology), and on "green chemistry", through the design of artificial protein catalysts.

Bioengineering 331, Stanford (Winter 2006-7)

Computational inference of biological regulation

Although the genomes of many organisms have been sequenced, these sequences are of little use without an understanding of the higher-level organization that controls the embedded genes. This research area (also known as computational systems biology) seeks to determine the structure of the systems which regulate the activity levels of genes and their products in order to produce biological function. Further understanding of this field would have effects not only on our understanding of biology, but also on medicine and pharmacology (by granting better understanding of the mechanisms of disease) and on synthetic biology (through a better understanding of the "architecture" behind existing biological systems).

Computer Science 279, Stanford (Autumn 2006)

Computer architecture

At Berkeley, two of my research projects were intimately tied in with architecture:

With Professor Bora Nikolic and Zhengya Zhang, I worked on architectures to allow efficient decoding of LDPC error-correcting codes. These codes come extremely close to the Shannon bound on code performance, but are NP-hard to decode in a maximum-likelihood fashion. Our work focused on efficient methods to implement belief-propagation-style decoding algorithms in hardware.

With Professor John Wawrzynek, I worked on custom architectures for molecular dynamics simulations on FPGAs, specifically on the BEE2 FPGA system.

Parallel programming languages

At Berkeley, I was part of the Titanium research group, working on a dialect of Java for high-performance parallel computing. My work (May-Dec 2004) involved library optimizations in Java. During my time at IBM Almaden, I also collaborated with Dan Bonachea on a port of Titanium to the Blue Gene/L supercomputer.

Computer network security

I was the Network Security Coordinator for two years (2003-2005) for UC Berkeley's Residential Computing. I did some research here in the use of benign worms for security and in the application of machine learning techniques to detection of compromised machines on a network.

Courses at Stanford

CS 229 - Machine Learning (Ng)
CS 279 - Computational Analysis and Reconstruction of Biological Networks (Koller)
Bioengineering 331 - Protein Engineering (Cochran)
CS 148 - Introduction to Computer Graphics (Hanrahan)
Structural Biology 241 - Biological Macromolecules (Herschlag et. al.)
Biochemistry 224 - Cell Biology of Physiological Processes (Theriot et. al.; audited)
ME 334 - Statistical Mechanics (Cai)