Close Collaborations Advance Progress in Genomic Research
A historical strength of much of the Laboratory’s science has been the ability to couple the development of predictive theoretical tools directly to experimental efforts that test the predictions. However, opportunities to exploit this highly productive approach have been rare in biology. The science is changing, and with it, the frequency with which this especially fruitful type of cross-discipline interaction occurs. One example of this interaction is illustrated in the article “Mining Genomes,” which discusses the work of bioinformaticist Ivan Ovcharenko working in collaboration with bioscientists Lisa Stubbs and Gabriela Loots.
The team’s work addresses the important but difficult challenge of identifying regulatory elements in genomes, which are small patches of DNA responsible for turning genes on or off as changing conditions require. To address this challenge, a suite of software tools is being developed to predict the location of these elements while several experimental approaches are used to test those predictions.
The tools developed by the Livermore team use a number of approaches to identify regulatory elements, but the central theme is an aspect of what biologists call comparative genomics. Here, the key concept is evolutionary conservation. Segments in genomes that encode important functions, such as the regulation of genes, are much more likely to be conserved over evolutionary time than is the rest of a species’ DNA. Thus, by comparing the genomes of different creatures, the conserved segments, including regulatory elements, can often be recognized.
By being “embedded” in our biological labs, Ovcharenko has the perfect environment to develop tools of maximum use to biologists. This close relationship ensures, among other things, a process through which tools can be continually honed and validated and at the same time made “biologist friendly.” One notable feature of Ovcharenko’s tools is the ease with which they can be learned and used by biologists who are unsophisticated in computer science. Indeed, much of what biological researchers need is not high-level computer programs running on supercomputers but personal computer programs written by people who understand biology and biologists.
Tools of the kind described in the article are at the forefront of genomic research; they are essential to extracting information from increasingly more extensive Web-accessible genomic data sets covering “all creatures great and small,” from bacteria to plants to humans. But they also reflect a profound shift occurring in biological research. That shift, oddly, is one of culture and belief, and of social standards—standards that dictate what approaches to research are acceptable. For a long time, the doctrine of hypothesis testing has dictated how biological science should be done. The idea is that experimental work should be undertaken to test a tightly defined hypothesis—one resting firmly on what is already well established. In stark contrast is a “fishing expedition” (now called discovery-based science) designed to gather information for its own sake in anticipation of what it will teach us. The Human Genome Project was a fishing expedition, and it led to an outpouring of wonderful discoveries that no one could have anticipated or framed as “hypotheses.”
Increasingly, biologists are becoming aggressive, free-ranging “hunter–gatherers” of the truth. They are using new high-throughput techniques to produce huge data sets and using computers and the Internet to search them en mass. It is in the contribution they make to this research revolution that the Livermore team’s tools have their true significance. They let us ask of our exploding collection of genomes, “What do you have to tell us?” The genomes’ answers are stunning, in part because they show us the limitation of adhering exclusively to hypothesis testing.
We are making exciting headway in unraveling the secrets of genomes and therefore of life itself. Obtaining a detailed picture of how genes and other DNA sequences function is an essential foundation to gaining a full understanding of how biological systems work and how they malfunction. The significance of this understanding will certainly be immense—in part because it will include a vastly deeper understanding of disease processes. This understanding will lead to better disease diagnosis, prevention, and treatment.