Eytan Ruppin and Keren Yizhak's work may lead to development of an anti-aging drug
The Department welcomes comments, suggestions and corrections. Send email to editor [at] cs [dot] umd [dot] edu.
Dr. Eytan Ruppin (incoming UMD Computer Science Department Professor and Director of the UMD Center for Bioinformatics and Computational Biology) and his Ph.D. student Keren Yizhak (Tel Aviv University, Israel) have developed an algorithm that predicts which metabolic genes need to be 'switched off' to prevent aging. Their work was motivated by previous research that has shown the efficacy of caloric restriction as a mechanism to increase life span in a variety of mammals, including humans. Dr. Ruppin and Ms. Yizhak sought to determine which combinations of genes should be targeted by drugs in order recreate these caloric restriction longevity effects on molecular and physiological levels. To this end, they developed a novel "metabolic transformation algorithm" (MTA) that can determine such genes of interest. So far, Dr. Ruppin and Ms. Yizhak have used the MTA to predict which genes in yeast should be suppressed in order to combat aging. Their experimental collaborators (the Haim Cohen lab in Bar-Ilan University) have successfully validated their predictions by suppressing the genes identified by the MTA. The suppression of these genes has more than doubled the longevity of the yeast.
Findings such as these, if validated in higher animals, could eventually lead to the development of revolutionary new drugs designed to slow aging. Dr. Ruppin plans to continue this work upon his arrival to the University of Maryland. He notes that "while we were pleased to see that the MTA algorithm that Keren has developed works well and yields valuable predictions, the latter were currently tested and experimentally validated only in yeast. Obviously there is a clear need to apply MTA to analyze aging data and predict life-extending genetic perturbations in higher animals. We indeed plan to generate such predictions now by analyzing aging mice data and then test those experimentally by our collaborators, and if all goes well, take it from there."