AI-Powered Platform Designs Peptides to Target Undruggable Proteins

Biomedical engineers at Duke University have developed an innovative AI-based platform called PepPrCLIP that designs peptides capable of binding to and destroying previously undruggable disease-causing proteins. This new approach, inspired by OpenAI's image generation model, offers a promising solution for targeting proteins that have been resistant to conventional therapies. 

The platform consists of two main components: PepPr, a generative algorithm trained on natural protein sequences to design new 'guide' proteins, and CLIP, an adapted version of OpenAI's algorithm that matches peptides with their targeted proteins.

Pranam Chatterjee, senior author on the study published in Science Advances, explains, “OpenAI’s CLIP algorithm connects language with an image. If you have text that says ‘dog,’ you should get an image of a dog,” said Chatterjee. “Instead of language and image, we trained it to match peptides and proteins. PepPr makes the peptides, and our adapted CLIP algorithm will screen those peptides and tell us which ones will make a good match.”

In comparative tests, PepPrCLIP outperformed existing platforms in speed and efficacy, generating peptides that were consistently better matches for their targeted proteins. The team conducted experimental tests on both ordered and disordered protein targets.

The platform successfully designed peptides that could bind to and inhibit UltraID, a stable enzyme protein, as well as beta-catenin, a disordered protein involved in cancer signaling. In their most challenging test, PepPrCLIP created peptides capable of binding to a highly disordered protein associated with synovial sarcoma, a rare and aggressive cancer 

Chatterjee emphasizes the significance of these results: "These complex, disordered proteins have made a lot of cancers and diseases practically undruggable because we couldn't design molecules that bind to them. But PepPrCLIP showed that it could work on even the most complicated protein, and that opens up a lot of exciting clinical possibilities."

The team plans to further improve the platform and collaborate with medical and industry professionals to develop peptides for potential therapies targeting diseases caused by unstable proteins, including Alexander's Disease and various cancers.