Pranam Chatterjee Receives NIH Award to Use AI to Target and Treat Disease

Pranam Chatterjee received a Maximizing Investigators’ Research Award (MIRA) from the National Institutes of Health. Given to NIH Outstanding Investigators, the award provides $1.25 million in funding over five years and is intended to support early-stage investigators as the pursue ambitious scientific projects.

Chatterjee, an assistant professor of biomedical engineering at Duke, will use the MIRA funding to develop generative AI algorithms to create molecules that can bind to––and potentially alter––proteins associated with cancer and other diseases.

“We’re trying to make CRISPR for proteins, and this grant gives us the flexibility to build out that system,” said Chatterjee. “With CRISPR systems, you have RNA that targets a stretch of DNA, and a Cas protein, which can bind to the DNA and cut it. We do the same thing, but we use AI to help us design a peptide that will target a key region of the protein’s amino acids. We then link the peptide to an enzyme that will modify the target.”

Because they don’t need to decipher the structure of the target protein, this approach allows Chatterjee and his team to aim for targets that have previously been out of reach. They are currently developing a platform to study how they can target and degrade the proteins associated with Alexander disease, an extremely rare genetic disease. But with the support of the MIRA program, Chatterjee and his team hope to expand this work to develop more precise targeting platforms and treat a wider array of diseases.

Their first goal involves creating new platforms that can create binding peptides that will only attach to the mutated forms of proteins.

“There is a protein called KRAS, and a single, small mutation to this protein is associated with several different types of cancer,” said Chatterjee. “Healthy forms of KRAS help with cell signaling, so you don’t want to knock out the non-mutated forms along with the mutated ones.”

To accomplish this, Chatterjee and his team will develop new AI platforms that can create peptides to specifically target and bind to the mutated protein targets. As they develop these algorithms, they’ll use experimental systems to test their molecules in the lab to identify which designs work best.

The team will employ a similar approach as they design platforms that can help target changes caused by incorrect chemical modifications rather than gene mutations.

After a protein is made, different chemicals will be attached to the protein to accomplish a variety of tasks that are not possible with the protein alone. But sometimes the wrong chemical will be attached to the protein, negatively affecting protein behavior and causing disease. One such example is a protein called STAT3. With the correct chemical additions, the protein is a key driver of cellular function, but an incorrect chemical modification alters the protein activity and causes breast cancer.

“STAT3 is present in every single cell,” says Chatterjee. “You don’t want to target and destroy them all. But if we can develop platforms that allow us to get more specific based on these mutations or chemical alterations, we can make therapies that are much safer and targeted. It opens up a big array of therapeutic opportunities.”

The team’s first goal involves the creation of “on-switches,” which are intended to help stabilize proteins. In previous work, Chatterjee and his lab developed therapeutic tools called “ubiquibodies,” which can bind to and degrade target proteins. With the support of the MIRA funding, the team now plans to create “deubiquibodies” that can help stabilize cells by preventing the destruction of key proteins.

“Sometimes cells will mark a protein as trash, even if it’s helpful for the body. For example, a cell could add a trash tag to a protein that’s actually a tumor suppressor,” said Chatterjee. “We know that you want that protein because it helps prevent cancer, so the deubiquibodies can remove that trash tag and make sure these helpful proteins are saved.”

It will be a tremendous amount of both computational and wet-lab work, but Chatterjee is looking forward to using his lab’s unique combination of skills as they explore how to enhance their AI-based platforms.

“If we’re successful, we could create tools that will allow us to precisely keep the proteins that help protect us and get rid of proteins that damage us,” said Chatterjee. “I’m grateful that the NIH has given us this opportunity to push this work forward.”