MIT Professor Develops Predictive Models to Understand Tumor Evolution

Computational approaches aim to decode the molecular processes driving cancer progression and drug resistance.

Published on Mar. 11, 2026

Matthew G. Jones, an assistant professor at MIT, is using artificial intelligence and machine learning tools to build predictive models that illuminate the complex evolution of cancer tumors. His research focuses on a specific mechanism called extrachromosomal DNA (ecDNA) amplification, which allows tumors to rapidly adapt and become more aggressive in response to treatment. By studying tumor lineages and evolutionary histories, Jones hopes to identify new therapeutic targets and strategies to overcome drug resistance.

Why it matters

Understanding the dynamic evolutionary processes that drive tumor progression is crucial for improving cancer treatment and patient outcomes. Tumors can evolve in unpredictable ways, often rendering therapies ineffective over time. Jones' work using computational approaches aims to decode the underlying molecular logic behind these transformations, which could lead to better ways to anticipate and overcome drug resistance.

The details

Jones' lab is focused on studying extrachromosomal DNA (ecDNA) amplifications, which are circular DNA fragments that exist separately from a cell's chromosomes. These ecDNA amplifications are found in about 25% of the most aggressive cancers, including brain, lung, and ovarian tumors. The ecDNA allows tumors to rapidly adapt and evolve in response to treatment pressures, accelerating disease progression. Using single-cell lineage tracing technologies, Jones and his team are mapping the evolutionary histories of tumors to understand how ecDNA drives these dynamic changes.

  • In the 1960s, ecDNA amplifications were first discovered in cancer cells.
  • In the 2010s, next-generation sequencing revealed ecDNA amplifications are far more prevalent than initially thought.

The players

Matthew G. Jones

An assistant professor in the MIT Department of Biology, the Koch Institute for Integrative Cancer Research, and the Institute for Medical Engineering and Science. He is using computational approaches to build predictive models of tumor evolution.

Koch Institute for Integrative Cancer Research

An MIT research institute that promotes collaboration between engineers and basic scientists to advance cancer research.

Got photos? Submit your photos here. ›

What they’re saying

“Tumors have an incredible, and very challenging, ability to evolve: the ability to change their genetic makeup, protein signaling composition, and cellular dynamics.”

— Matthew G. Jones, Assistant Professor (Mirage News)

“We're focused on one specific way tumors are evolving through a form of DNA amplification called extrachromosomal DNA. Excised from the chromosome, these ecDNAs are circularized and exist as their own separate pool of DNA particles in the nucleus.”

— Matthew G. Jones, Assistant Professor (Mirage News)

What’s next

Jones hopes his research will help stratify patients who will respond to certain drugs, anticipate and overcome drug resistance, and identify new therapeutic targets to improve patient outcomes.

The takeaway

By using computational approaches to decode the complex evolutionary processes driving tumor progression, Jones and his team aim to develop better ways to stay ahead of cancer's ability to adapt and resist treatment, ultimately leading to more effective and personalized cancer therapies.