Scientists at UC San Diego have developed a new AI platform called POLYGON that has the potential to revolutionize drug discovery. Unlike existing protocols, POLYGON can identify molecules with multiple targets, which is significant as multi-target drugs have the potential to deliver the same benefits as combination therapy, but with fewer side effects. To train POLYGON, researchers used a database of over a million known bioactive molecules, allowing it to generate new candidate drugs with specific properties. They synthesized 32 molecules that had strong predicted interactions with cancer-related proteins MEK1 and mTOR, suggesting they could potentially target both proteins as a cancer treatment.