About MDD - Subscription Info
December 2001
Vol. 4, No. 12, pp 15.
news in brief
Function from fusion
figure: proteomic split ends
Proteomic split ends. Researchers have found that similarities between parts of a single protein from one organism (left) to different proteins in a second organism might just signal a functional association between the proteins of the second organism.
Although the fruit fly is superficially distinct from yeast and humans, we are beginning to see how truly similar these organisms are at the molecular level. Initial comparisons made at the DNA level looked for simple homologies between the genes of one organism and those of another, but recently, researchers have started to realize that the homologies may extend beyond individual molecules to metabolic complexes. And it might not just be a one-for-one deal. Instead, where two proteins perform a given function in one organism, one protein—bearing similarities to both of its kin—might be sufficient in another. This concept, gene fusion, is common throughout molecular evolution and can be advantageous to the cell by reducing the regulatory machinery needed for a particular function.

Anton Enright and Christos Ouzounis of the European Bioinformatic Institute (Cambridge, UK), however, have turned this concept around (Genome Biol. 2001, 2, 1–7). “The detection of gene fusions in one genome (defined as ‘composite’ proteins),” they write, “allows the prediction of functional associations between homologous genes that remain separate in another genome (defined as ‘component’ proteins).” The researchers developed an algorithm that detects just such homology patterns between genes in various genomes, in the hopes of defining new interactions between otherwise disparate proteins, including those with unknown functions.

The scientists applied their algorithm to the genomes of 24 organisms, sequentially comparing one genome to the other 23 and finding 7224 component proteins that identified with 2365 composite unique proteins, results that could be tested using functional genomics.

A particular functional association that the algorithm predicted was between the yeast proteins MXR1, a peptide methionine sulfoxide reductase involved in antioxidative processes, and YCL033C, a protein of unknown function. Together, the proteins appear to be part of a process to protect the yeast from oxidative damage, which highlights this approach’s utility for elucidating function.

Thus, the algorithm should help researchers identify new proteinaceous partners within various organisms, thereby expanding the number of potential drug targets.

RANDALL C. WILLIS

< Previous Article

Return to Top || Table of Contents

 CASChemPortchemistry.orgPubs Page