In an effort to detect cancer at its earliest stages and reduce cancer-related deaths, Indian American Massachusetts Institute of Technology (MIT) researcher Sangeeta Bhatia, working with Microsoft researcher Ava Amini and colleagues, is using artificial intelligence to design molecular sensors for early cancer detection.

According to MIT News published on January 6, Bhatia and Amini ’16, a principal researcher at Microsoft Research and a former graduate student in Bhatia’s lab, are the senior authors of the study, which was published on January 6 in Nature Communications. The paper’s lead authors are Carmen Martin-Alonso PhD ’23, a founding scientist at Amplifyer Bio, and Sarah Alamdari, a senior applied scientist at Microsoft Research.

The researchers developed an AI model to design peptides, or short proteins, that are targeted by enzymes known as proteases, which are often overactive in cancer cells. MIT News noted that nanoparticles coated with these peptides can act as sensors, emitting a signal when cancer-linked proteases are present anywhere in the body.

Depending on which proteases are detected, physicians could determine the specific type of cancer involved. These signals could be identified using a simple urine test that could potentially be performed at home, according to MIT News.

“We’re focused on ultra-sensitive detection in diseases like the early stages of cancer, when the tumor burden is small, or early on in recurrence after surgery,” said Bhatia, the John and Dorothy Wilson Professor of Health Sciences and Technology and of Electrical Engineering and Computer Science at MIT, and a member of MIT’s Koch Institute for Integrative Cancer Research and the Institute for Medical Engineering and Science.

MIT News noted that more than a decade ago, Bhatia’s lab introduced the concept of using protease activity as an early marker of cancer. “The human genome encodes about 600 proteases, which are enzymes that can cut through other proteins, including structural proteins such as collagen. They are often overactive in cancer cells, as they help the cells escape their original locations by cutting through proteins of the extracellular matrix, which normally holds cells in place,” the report stated.

The researchers proposed coating nanoparticles with peptides that can be cleaved by specific proteases. These particles could be ingested or inhaled and, as they traveled through the body, would encounter cancer-linked proteases that cleave the peptides, MIT News explained.

The cleaved peptides would then be excreted in the urine, where they could be detected using a paper strip similar to a pregnancy test. Measuring these signals would reveal abnormal protease activity deep within the body.

“We have been advancing the idea that if you can make a sensor out of these proteases and multiplex them, then you could find signatures of where these proteases were active in diseases. And since the peptide cleavage is an enzymatic process, it can really amplify a signal,” Bhatia said.

The research team has previously demonstrated diagnostic sensors for lung, ovarian, and colon cancers. However, MIT News noted that earlier studies relied on a trial-and-error process to identify peptides cleaved by certain proteases. In many cases, those peptides could be cleaved by more than one protease, making it difficult to attribute signals to a specific enzyme.

Even so, using multiplexed arrays of multiple peptides produced distinctive sensor signatures that proved diagnostic in animal models across several cancer types, even when the exact proteases responsible for cleavage were not clearly identified.

In the new study, the researchers moved beyond this traditional approach by developing a novel AI system called CleaveNet, designed to generate peptide sequences that can be efficiently and specifically cleaved by targeted proteases.

Users can prompt CleaveNet with specific design criteria, and the system generates candidate peptides likely to meet those requirements. This allows researchers to fine-tune both the efficiency and specificity of the peptides, improving the diagnostic power of the sensors.

“If we know that a particular protease is really key to a certain cancer, and we can optimize the sensor to be highly sensitive and specific to that protease, then that gives us a great diagnostic signal,” Amini said. “We can leverage the power of computation to try to specifically optimize for these efficiency and selectivity metrics.”

For a peptide composed of 10 amino acids, there are roughly 10 trillion possible combinations. Using AI to navigate this vast search space enables faster prediction, testing, and identification of effective sequences, while significantly reducing experimental time and cost.

The research received funding from the La Caixa Foundation, the Ludwig Center at MIT, and the Marble Center for Cancer Nanomedicine.

Related posts:

Microsoft’s $3 Billion Investment in India to Boost AI Innovation and Skilling Initiatives Indian-American Community Honors Anantha Chandrakasan, First Indian-American Provost of MIT “Grey’s Anatomy” Star Patrick Dempsey Advocates for Insurance Coverage of Holistic Cancer Therapies  Reliance Foundation and UN India Host ‘Early Warning to Early Action’ Convention in Odisha