Skip to main content
Premium Trial:

Request an Annual Quote

Promises and Problems

Accepting the D.C. White Research and Mentoring Award at ASM this year, the J. Craig Venter Institute's Ken Nealson said the promises made by genomics and metagenomics have served to both draw him in and expose the problems inherent in both fields. The first promise, Nealson said, is that learning an organism's functions and the structure of its genome will enable researchers to predict how it will act in nature. Nealson used his work with the bacteria Shewanella oneidensis to illustrate problems with the first promise. Many genes don't do what they're supposed to do, he said. And in different samples of the bacteria, some functions previously seen in other experiments exist without the requisite genes, while in other cases the genes exist without the previously-seen functions. Another problem, Nealson said, is that about half the microbe's genes are almost impossible to identify, something he likened to "playing poker with only half a deck." The problems won't go away, he said, and will likely be a part of any microbe that is studied. But, by using comparative genomics and gene function validations, it may be possible to solve the issue, "one gene at a time." He added, "It's the 'dark matter' that's going to keep us from understanding what's going on. It's tough, painstaking work, but it has to be done." The second promise is that genomic analysis will lead to an understanding of organisms and their functions. But the problem with this, Nealson said, is that there's simply too much data. "Sequencing is almost free these days and we're swimming in a pool of data," he added. This became obvious with Shewanella when Nealson and his colleagues saw there were at least 20 strains of the bacteria, all with their own variable characteristics. The third promise, Nealson continued, is the closest to being right: metagenomic analysis of an environment can lead to understanding organisms and their functions. But the key to greater understanding, he said, is to reduce the amount of "dark matter" in genomics, to keep doing the time-consuming work that shines a light on the unknown.