My primary interest lies in developing algorithms to analyze massive or otherwise intractable data sets produced by molecular biologists. In my thesis work, I applied randomized strategies to efficiently automate large-scale tasks in genome annotation, such as discovering repeats in genomic sequence (such as the Drosophila genome shown above), comparing very long orthologous sequences, and finding subtle sequence motifs.
More specifically, I use the technique of random projection onto subspaces (a.k.a "locality-sensitive hashing") to reduce certain inexact matching tasks to a series of exact matching problems that can be solved more efficiently. You can read about my work in Bioinformatics (for genome comparison) and RECOMB 2001 (for motif finding). For more information, see the references in the publications section of my CV.
Recently, I acquired a new computing cluster to run big sequence comparison jobs.
If you're looking for a reprint of the Bioinformatics article, here is my personal version (posted 5/2/2001): PDF Postscript. (The journal's version has nicer formatting but should have identical content).
I also occasionally hack on Dapple, a program for image analysis of DNA microarrays.
Finally, you might have come here looking for my tutorial page on cDNA microarrays and comparative hybridization.