New Haven, Conn. — Scientists at Yale and in Singapore have devised what may be the ultimate acid test — a comprehensive model for estimating the origins of Earth’s habitability, based in part on ocean acidity.

The new theoretical model applies previously published, Yale-led research to a wide range of interconnected geological and atmospheric processes. It may provide the clearest picture yet of how Earth evolved to a point where life was able to flourish.

A team of researchers from USC has unlocked the details of a cellular pathway that triggers cellular inflammation and aging and is linked to Alzheimer’s disease, particularly among those who carry the APOE4 genetic risk. They have also found a way to return cells to a healthy state, revealing a new potential approach to treatment. Past research found that a shortage of HDL cholesterol in the brain raises a person’s risk for Alzheimer’s disease. That risk is related to problems with ABCA1, which produces HDL when working properly. To discover what is going wrong with ABCA1 and how it increases risk, scientists used a range of methods to pinpoint the processes unfolding inside brain cells. They found that in brains of people affected by Alzheimer’s disease or who carried the APOE4 gene putting them at higher risk for the disease, ABCA1 increased, but became trapped in a part of the cell that typically clears waste. That change was linked to a rise in a modified form of cholesterol known as oxysterol. Lowering oxysterol, in both animal models and human stem cells, freed the trapped ABCA1 and restored the pathway to its healthy state. Lowering oxysterol could be a new way to prevent or treat Alzheimer’s disease in its earliest stages.

Lehigh University researchers are pioneering AI-driven advancements in materials science with an $800,000 grant from the U.S. Department of Energy. Led by Dr. Chinedu Ekuma, the project focuses on developing interpretable machine learning models to predict material properties more efficiently. Key innovations include physics-guided machine learning, scalable AI for material predictions, integration with diffusion models, and an open-source computing platform for global accessibility. This research aims to accelerate breakthroughs in semiconductors, aerospace materials, battery technology, climate modeling, and healthcare.