Skip to main content
Premium Trial:

Request an Annual Quote

Kimmen the Driven

Premium

At 17 she dropped out of college. At 45 she coauthored Celera’s human genome paper.

by Ken Howard

 

If ever there was a triumph-over-adversity story for the bioinformatics set, Kimmen Sjölander’s is it. The principal scientist in Celera Genomics’ Foster City, Calif., protein informatics group, whose CV now includes “The Sequence of the Human Genome, Science, 2001 Feb 16,” seems genuinely thrilled by the string of career successes and personal blessings she has experienced over the past few years. And she is proud to tell you what she overcame to achieve them.

Since 1996, Sjölander has been wowing her peers by developing algorithms that cluster proteins to predict their function and structure — or, as she puts it, to “elucidate evolutionary history of molecules ... and see how the molecules have changed to help determine function.” Watching her present her work to a roomful of the world’s leading computational biologists, it’s hard to believe that her bioinformatics career took root when, as a college dropout and divorced mother of three, she enrolled in a community college computer science course.

Today, Sjölander’s credits include a US patent, issued last October, for a Bayesian sequence analysis method and apparatus, as well as the Panther technology, which is now integral to Celera’s discovery software and in-house datamining efforts. Panther’s genesis can be traced back through her career: it took root while she was a doctoral candidate under David Haussler at the University of California at Santa Cruz, and evolved with her through senior scientist posts at Pangea Systems (now DoubleTwist) and Molecular Applications Group (now defunct) before Celera acquired the tool and her team in 1999.

Haussler recalls, “She was a real powerhouse at Santa Cruz. We recruited her as an undergraduate here in the early days of hidden Markov models and super family classification.”

Sjölander says Haussler “approached me to work with him on methods for multiple sequence alignment using hidden Markov modeling. It was a first time for [using] hidden Markov for proteins, so I was on one of the first papers to come out and it snowballed.”

Vim and rigor

Sjölander was chosen to represent UCSC at the second Critical Assessment of Structure Prediction contest in which teams receive protein targets with crystallization results not yet announced. “This really separated those blowing hot air from those who could really predict things,” says Haussler. The contest is “contentious and exciting like the Academy Awards. It was a great moment for seeing our work,” he adds.

For Sjölander, whose team was a finalist, it was also a signal event. It affirmed for her the skills she had developed for applying Bayesian evolutionary tree estimation, Dirichlet mixtures, and hidden Markov models to successful protein prediction. “Kimmen had really concentrated on a lot of important questions in studying proteins,” says Stephen Altschul, a BLAST author and senior investigator at the National Center for Biotechnology Information.

Altschul says Sjölander’s Dirichlet mixture method provided a rigorous mathematical framework for constructing models of protein families from a small number of members of families. “It’s the only method that has a really sound mathematical foundation, so it’s a nice way to think about the problem,” he says. “If you have a good theory behind a method it gives you a logical way to refine it — you’re not shooting in the dark as much.”

The creative life

The speech she delivered at CASP not only capped her success as a graduate researcher and launched her career, but also served to let Sjölander know she had “made it” in science, a world she’d been exposed to early on.

Sjölander grew up in a United Nations village in New York City among families of Japanese businessmen, UN staffers, and “left-wing others,” the group to which she says her siblings and her mother, a historian of physicists, belonged. “We had a lot of physicists visit the house. Growing up in that environment influenced what I went into. You get a vision of what science can be like — a very creative and interesting life.”

Sjölander took an indirect route to realizing the scientific life for herself. She entered City University of New York at 15, but dropped out at 17 and went off to San Francisco, escaping the confines of the one-bedroom apartment that she had shared with her mother, brother, and identical twin sister.

In California, Sjölander pursued a free-spirit artist’s lifestyle doing sculpture, painting, and writing, performing her own music, studying with an Indian guru, and working in Carlos Santana’s restaurant. She also married and had three children.

When her marriage collapsed irreparably, Sjölander summoned the courage to take her kids, get out, and go back to college. The transition was difficult, she recalls. “I was doing research right away and undergoing custody battles. I got straight As but it was difficult with three small children.”

Smack-dab driven

Her accomplishments are no surprise to Haussler, who says she has “pure drive.” At UCSC, Sjölander “made connections between computation and biochemistry groups,” he says. “She’s very aggressive in building an interdisciplinary approach and we’ve programized her approach that epitomized her graduate study here.”

The approach also defines her as a scientist. Says her twin, Tandy Warnow, who codirects the Center for Computational Biology and Bioinformatics at the University of Texas at Austin and whom colleagues understandably often confuse for Sjölander at conferences: “She’s one of the most true smack-dab-in-the-middle bioinformaticists I know. She’s not primarily a computer scientist or biologist, she’s really a bioinformaticist.”

Sjölander offers a more explicit classification: “I’m a computational molecular biologist. Bioinformaticists are, in some people’s minds, people who can go on the web and use tools like BLAST. I am a hybrid kind of scientist who creates tools, but on the other hand I’m also a tool user using data for biological discovery.”

Still, she’s not done evolving. While a recent development in her personal life has her heart singing — she and nuclear physicist and Informix engineer Nigel Godwin were married in Hawaii last year — Sjölander lists a few more career ambitions: “I’d like to be more and more involved in the discovery side. I typically make tools, but discovery takes part in using them. I also want [a role in] finding keys to disease, uncovering pathways and figuring out why we are what we are.”

 

The Scan

NFTs for Genome Sharing

Nature News writes that non-fungible tokens could be a way for people to profit from sharing genomic data.

Wastewater Warning System

Time magazine writes that cities and college campuses are monitoring sewage for SARS-CoV-2, an approach officials hope lasts beyond COVID-19.

Networks to Boost Surveillance

Scientific American writes that new organizations and networks aim to improve the ability of developing countries to conduct SARS-CoV-2 genomic surveillance.

Genome Biology Papers on Gastric Cancer Epimutations, BUTTERFLY, GUNC Tool

In Genome Biology this week: recurrent epigenetic mutations in gastric cancer, correction tool for unique molecular identifier-based assays, and more.