Communications of the ACM
Network Analysis Predicts Drug Side Effects
After linking drugs (on the left) to their known side effects (on the right) in a network, researchers trained a computer to predict new connections, an approach that might make drugs safer before they hit the market.
Credit: Cami et al/Science Translational Medicine 2011
Researchers at Harvard Medical School and Children's Hospital Boston, led by Ben Reis and Aurel Cami, have developed a mathematical network to predict drug side effects that normally are not discovered until thousands of people have taken the medication.
The researchers started with a 2005 catalog of existing medications and their known side effects. They linked the drugs and their side effects in the network, and then created a program to predict likely new connections between drugs and side effects. The system was able to predict 42 percent of the drug-side effect relationships that were later found in patients, according to the researchers.
The network included information on chemical properties, such as the drug's melting point and molecular weight, and where the drug acts in the body. The program also was able to use this data and relationships to predict side effects that were reported years later.
The researchers are now studying what kinds of data work best and trying to determine drug interactions that also can be dangerous but are rarely studied in clinical trials. "We’re moving from a paradigm of detection--where it takes sick people to know something is wrong--to prediction," Reis says.
From ScienceNews
View Full Article
Abstracts Copyright © 2011 Information Inc. 
, Bethesda, Maryland, USA
No entries found