Because of time and resource constraints, however, researchers cannot directly study in detail the genomes of all of the thousands of species of bacteria and can only carry out experiments on some of them, Sze explained. So researchers need a computational tool to help them predict where similar clusters of genes are in different species of bacteria so they can better focus their experiments, he said.
For each level of complexity in organisms, there is a model organism that scientists center their experiments on, and for bacteria, the model organism is E. coli. Because E. coli is a model organism, scientists have studied it in great detail and have a large amount of experimentally validated data on its genome and the operons that function, Sze said.
Sze and Yang's computational tool starts with a known and experimentally validated E. coli operon and then searches each of hundreds of separate species of bacteria for genes that are related to those in the E. coli operon. Once the tool has located related genes in a bacterium, it then checks to see if there is a strong clustering of the genes. It does this by using a statistical procedure that computes which genes are statistically very close to each other when you compare it to a random situation of genes, Sze explained.
"Imagine putting genes in a random order," Sze said. "When placed randomly, two specific genes are likely to be located at a far distance from each other. But if there is clustering, then the genes will be much closer to each other than they would be by chance."
So by using Sze and Yang's tool, scientists can easily locate sub-blocks of genes in different bacteria that are arranged in a similar fashion as one of the operons in E. coli. "If biologists are interested in a particular E. coli operon, they can use our tool to find where the operon is in the different bacteria," Sze
|Contact: Sing-Hoi Sze|
Texas A&M University