|
02/14/2007, 2:30 PM - 3:30 PM
Speaker:
Daming Li, Executive Vice President
, LITEC Systems Corporation
.
Gene Expression Programming (GEP) and Parallel Genetic Algorithm (PGA) on Linux grid computing cluster have proven to be a very effective way of tackling intensive mathematical problems in financial modeling and drug discovery.
Based on the features of stock objects, we present our GEP grid computing models including the fitness that appropriated to the special rules of stocks, give experiments and analysis on the real stock price index of NYSE. The results show that the precision that predicts by using our models is higher than traditional method.
The availability of molecular structures of drug targets and candidate compounds has opened the door for the application of large scale grid computing technology to conduct virtual drug design. Through the use of PGA on Linux grid computing cluster, we developed the computing power to effectively discover potential drug candidates. Our models generate better profiles of combinatorial drug candidates optimized by PGA.
|
