Kimmen Sjolander's Home Page

Kimmen Sjolander's Home Page

summer, '95. From a machine in a Paris metro station, my mugshot for the monthly metro pass, Coupon Orange...

NEWS! I've taken a faculty position at the Department of Bioengineering at UC Berkeley, where I will be working on my favorite research problems in structural and functional biology: remote homology detection (and fold prediction), identification of protein domains, predicting binding pockets in proteins, subfamily classification, molecular evolution, protein-protein interaction, and so on. As a new faculty member, I'll be very interested to find people who are interested in this kind of work, who'd like to come to Berkeley for their doctoral work or for postdoctoral positions. Write to me!

Now, on to the really important things: pictures of my monkeys...

I got my name from my Swedish-American mother's enthusiasm for classical Greek literature. Funny thing, though: she changed the spelling of Cimon, (a rather interesting character from Plutarch's Lives) to Kimmen, giving me an authentic Swedish name -- a nickname for men named Joakim! My guess is she heard this name from her parents, who would speak Swedish when they didn't want the kids to understand what they were saying...

From 1999 to November, 2001, I worked at Celera Genomics, and was privileged to have helped the annotation of the newly sequenced human genome, published in Science.

Unforunately, we missed a couple of errors in the citations before publication ;-(.

Personal Data:

I finished my PhD at UCSC, under David Haussler. My thesis involved the statistical inference of evolutionary relationships among proteins, using a variety of different tools from computational learning theory. See our ftp site for various papers, including my Ph.D. thesis. Apart from this interest in evolution, my main research interests have been on statistical models for interesting problems in molecular biology, such as computing the expected distributions of amino acids in a position of a protein using Dirichlet mixtures, protein structure prediction , inter-molecular interactions (as in protein-dna binding sites), inferring and incorporating phylogenetic relationships among protein sequences into hidden Markov model (HMM) construction, and so on.

See compbio papers and my curriculum vita for some of the work I've been involved with over the last couple of years.

Pre-print of ISMB98 paper --an analysis of SH2 domains demonstrating a method for inferring phylogenetic trees using a combination of Bayesian and Information-theoretic tools, and identifying subfamilies. This analysis produced a new subfamily classification in SwissProt for Src2_drome, and proposes a relationship between the SH2 domains of Nck, Grb2, Sem5 and Drk.

Here at Santa Cruz, our computational biology group has been doing a lot of interesting work (okay, I'm biased). Check out our compbio home page.

Last modified: February 14, 2001

Kimmen Sjolander / kimmen@alumni.cse.ucsc.edu