A new paper published in Nature, Large-scale prediction and testing of drug activity on side-effect targets, details how a computer program was used to test 665 marketed drugs on 73 unintended ‘side effect’ targets. Since drugs frequently interact with more than one target, it is necessary to test new drugs for adverse side-effects before they are put into routine use. Many times adverse side effects are not seen until the drug has been on the market for several years, thereby endangering many users for unknown drug complications. With this new computer program, new drugs can be modeled to determine possible side-effects, which could save millions or even billions of dollars by preventing harmful drugs from ever going to clinical trials.
As an example of data collected, an interesting finding with this study was that a known drug side-effect for a synthesized form of estrogen, stomach pain, did not have a known target for this side-effect. The computer model suggests this synthesized estrogen strongly binds to COX-1, which is also a target for non-steroidal anti-inflammatory drugs, such as aspirin, which also can cause stomach pain, ulceration, and bleeding.
For more information, you can also visit Science Daily.