AI models can generate more complete summaries of complex cancer pathology reports than physicians, according to a new Northwestern Medicine study that tested six models developed by Meta, Google, DeepSeek and Mistral AI.

Myelodysplastic syndromes (MDS) are malignant disorders arising from hematopoietic stem cells, yet the precise epigenetic abnormalities occurring in the primitive stem cells remain poorly understood. A collaborative team led by Professor Xuetao Cao at the Chinese Academy of Medical Sciences has generated the base-resolution DNA methylation map of human MDS hematopoietic stem cells. The study reveals the distinct patterns of methylation in MDS stem cells, offering insights into MDS pathogenesis and potential therapeutic targets.

A new study from the Max Planck Institute of Immunobiology and Epigenetics in Freiburg offers a likely explanation for why BET inhibitor cancer drugs have repeatedly underperformed in clinical trials: they treat two distinct proteins, BRD2 and BRD4, as if they do the same job. They don't. BRD4 drives the step most drugs target — pushing genes into active transcription — while BRD2 acts earlier, assembling the molecular machinery that gets the process started. Blocking both simultaneously creates unpredictable, hard-to-manage effects. The findings suggest that future drugs designed to distinguish between these two roles could lead to more targeted and effective cancer treatments.

The study reveals that 70% of the cellular response to metabolic stress is regulated at the translation of messenger RNA—a level of control that redefines how we interpret the gene expression reprogramming associated with liver disease.

The protein CPEB4 binds to these messenger RNAs and regulates the cellular response to stress by controlling the balance between signals that activate or limit the production of the proteins encoded by these RNAs.

Published in Science Advances, the work identifies a new mechanism regulating translation that is key to the liver's response to metabolic stress.

POSTECH reveals how mosaic partial reprogramming enhances wound repair responses.