Computer Program Can Identify Side Effects from Drugs in Early Development

A computer program developed by researchers at the University of Texas can predict whether drugs in the early stages of development will have harsh side effects.

This software assesses the chemical structure of a drug molecule, allowing researchers to determine whether these key sub-structures are in molecules known for aiding side effects in other drugs.

Although there has been other less successful software, this new program was tested on nearly 900 drugs against a database of 1385 side effects, demonstrating its superiority over prior software.

By using this system, it could help alert regulatory authorities and health care workers on what side effects could occur when a new drug enters the late stage of a clinical trial and then brought to the market.

Now, the software has been used to test compounds within the DrugBank database.

Once the software was used to screen 2883 small molecule uncharacterized drugs, researchers had the ability to predict a vast array of side effects, especially those that were missed by other methods.

“Based on drug side effect information from different sources, we verify that the prediction by our approach is indeed correct,” wrote researchers Jamiul Jahid, MD and Jianhua Ruan.

Each year in the United States, more than 2 million people experience adverse reactions to their medications. Furthermore, approximately 100,000 people die from problems associated with these side effects.

Although a majority of these people have serious illnesses and there are many factors that contribute to their deaths — meaning the side effects are not completely responsible – this software has the potential to flag problems ahead of time and ultimately reduce that number.

Additionally, these side effects could be used to help medicinal chemists better understand the underlying mechanisms that side effects arise from. In the future, this could allow them to design out that particular character from the drug molecules, in turn reducing the amount of drugs that go through the R&D pipeline that then fail in clinical trials because of side effects.