Steven Salzberg is a big-picture biologist trapped in a computer scientist's body.

As a graduate student in computer science at Harvard University, he read about this new thing called the Human Genome Project. Convinced that this would be where the action was for some time to come, he attended biology classes and read textbooks to get up to speed on the unfamiliar vernacular of biology. As an assistant professor at Johns Hopkins University, he started working on computational techniques for understanding DNA. (see CV)

But, says Salzberg, coming to The Institute for Genomics Research (TIGR) in Rockville, Maryland, was his most pivotal career move because it was here that he became truly immersed in genomics. Most importantly, he became aware of problems that his colleagues experienced on a daily basis that he could address through his work.

Since his arrival, his TIGR research group has developed 20 major systems to assist in locating genes, genome assembly and sequence alignment. His efforts have helped to bring bioinformatics to the fore of genomics research. He has made all of the programs open-source and thus freely available on the web. “If I make it open source, others can use that and the field moves ahead faster,” he explains.

In the past, he says, learning enough molecular biology to be effective in bioinformatics was his most daunting task. But turning the new Center for Bioinformatics and Computational Biology at the University of Maryland into a world-class centre is his next big career challenge. He will take over as its director in July and is keen to ensure that bioinformatics stays closely tied to genomics. “Sequencing centres are churning out genomes at an incredible rate, so comparative genomics will be an active area for some time to come,” he says.

To have the most impact, Salzberg encourages his graduate students to tackle projects that a lot of people care about. He eschews work based solely on the current technological challenges in computer science. “I discourage my students from working on problems that only 50 other people in the world understand,” he says.

And, not surprisingly, he presses his students to seek answers through collaborations with scientists in other disciplines. When it comes to unlocking secrets buried deep in genomes and revealing evolutionary lineages through comparative genomics, this computer scientist is adamant that to blaze a trail you need a blatant disregard for disciplinary boundaries.