AI co-scientist proposes condensate strategy to target MYC in cancer

September 21, 2026 at 07:33 PM2 min

This digest was compiled by AI from multiple sources — links to the originals are below.

An AI system called Co-Scientist generated 108 strategies for targeting the cancer-driving protein MYC and rejected all but one. The surviving strategy reverses the laboratory's previous thinking on condensates. The system reviewed more than 700 papers overnight after researchers corrected its initial misconceptions.

Key Facts

Co-Scientist reviewed more than 700 scientific papers and generated 108 possible strategies for targeting MYC.
The AI system rejected all but one of the 108 strategies as unworkable.
Anna Pertl and Kalon Overholt spent close to an hour correcting Co-Scientist's misconceptions before it understood the question.
Richard Young's 2018 research showed that cells switch on crucial genes by gathering regulatory proteins into condensates at super-enhancers.

The AI System

Co-Scientist launches several autonomous AI agents to search for and synthesize information from papers, evaluate competing explanations, refine and critique hypotheses, and iteratively test ideas against published evidence. The process can involve vast quantities of computational power because agents pursue distinct lines of reasoning before converging on workable solutions. Anna Pertl, a biochemist at the Whitehead Institute for Biomedical Research in Cambridge, Massachusetts, used the system on a rainy Tuesday in July.

The MYC Challenge

Pertl asked Co-Scientist for non-obvious but practical ways of harnessing the biology of molecular droplets known as condensates to shut down MYC, a protein that runs amok in most cancers. The request built on years of research by Pertl's supervisor, Whitehead biologist Richard Young, who in 2018 found that cells switch on crucial genes by gathering regulatory proteins into condensates at super-enhancers. Young also worked out how cancer cells hijack these super-enhancers to send the gene encoding MYC into overdrive, a strong hint that condensates help to fuel the high expression rate of the protein. Dewpoint Therapeutics in Boston, Massachusetts, a company Young co-founded, is now exploring direct targeting of tumour cell condensates.

Correcting Misconceptions

Pertl and Whitehead bioengineer Kalon Overholt spent close to an hour going back and forth with Co-Scientist before it understood the question. At first, the AI model assumed that the protein clusters at the centre of the research were meant to be the drug rather than the target. The tool also assumed that the clusters always activated gene expression, when in fact some do the opposite. Once Pertl and Overholt corrected these misconceptions, they set the system loose, and by the next day Co-Scientist had combed through more than 700 papers and generated 108 possible strategies.