PALO ALTO, Calif.—The difference between a salve and a poison can be simply a question of dose. Many drugs, such as the anticoagulant warfarin, have a very narrow therapeutic index, meaning too much or too little of the drug can lead to disaster for the patient. Sometimes, this challenge is exacerbated by large effective-dose variations between patients.

 

Researchers at Stanford University and Roche Bioscience approached this problem with computational pharmacogenetics in mice, presenting their work in Nature Biotechnology.

 

Initially, the researchers examined the processing of warfarin in mice to identify individual metabolic variations. They then performed computational haplotype analysis of specific P450 enzymes to look for correlations with the metabolic characteristics of the mice. The researchers found strong correlations between mouse strains that linked genetic legacy with drug metabolism. They confirmed their computational findings experimentally, noting that the degree of warfarin metabolism was directly linked to the amount of CYP2c29 P450 enzyme.

 

"This research and the computational method can help scientists and clinicians better understand the drugs they are using, as well as the diseases they target," stated Dr. Gary Peltz, head of genetics and genomics at Roche. "It can also be used to identify genetic susceptibility factors affecting drug-induced toxicity."