Traditional drug discovery efforts have evolved from a “trial and error” approach to one that is more scientific and rational. This is due in large part to our increasing understanding of chemical and biological processes and the availability of advanced computational resources. Now, more researchers are relying on computer models in many aspects of the drug discovery process, from identifying potentially useful compounds to making predictions about how the drugs may work in the body—and whether they could have unexpected side effects.

This month, an article in The Scientist highlighted a number of computational tools that are readily available and widely used by researchers focused on drug discovery. These tools allow researchers to virtually screen a library of compounds to identify those that may bind their target protein of interest, identify drugs that are similar in shape and chemistry to their compound of interest, and make predictions about whether the compound will take part in interactions which could lead to adverse drug effects. Such computational tools will improve our current drug screening process and can extend into other areas of research as well.

While the Food and Drug Administration currently requires animal testing in the drug review process, in silico tools have the potential to significantly reduce animal testing compared with more traditional approaches, as drugs that lack sufficient pharmacological activity can be removed from the approval pipeline before animal testing is involved, and compounds of interest can be prioritized for testing.

What do you think of the use of computer models in the drug discovery process? Please send your questions and comments to sciencecorner@navs.org. I look forward to hearing from you.

–Dr. Pam Osenkowski, Director of Science Programs

For more information see: The Scientist