Research can be a frustrating endeavor. What works well theoretically or in the lab may fail miserably when put into practice. Scientists often face this problem when searching for new battery materials. Yet a recent research paper has shown how AI can alleviate this problem by guiding researchers to the correct experiments to perform.

Controlling electron spin could open new possibilities for more efficient catalysts.  By reviewing recent research, scientists explore how magnetic fields affect catalyst performance and what challenges remain before these effects can be put to practical use. 

Making better use of valuable metals could help create more efficient catalysts. A new method allows more ruthenium to be incorporated into single-atom catalysts without compromising their stability.

Researchers at Kanazawa University and the National Institute of Genetics in Japan have used high-speed atomic force microscopy to observe how a group of proteins known as Smc5/6 interacts with DNA. The images reveal how this structure opens and closes, captures DNA and brings separate sections together. The researchers also observed Smc5/6 holding twisted sections of DNA in place, producing more tightly packed arrangements. The findings provide new insights into the molecular processes that contribute to the maintenance and repair of genetic material.

Heat generates power, but it also generates waste. Thermoelectric materials help recover energy that would otherwise be lost as heat. Researchers at Tohoku University have unveiled such a material that delivers a 250-fold increase in power simply by adding a small amount of titanium to chromium nitride.

Applying insights from animal experiments to our understanding of humans requires knowing how gene functions correspond across species. DNA sequence similarity alone, however, does not always reveal these relationships. To address this challenge, a research team led by Vice Director/Principal Investigator Yasuaki Hiraoka, Researcher Yuya Tokuta, and Associate Professor Yusuke Imoto at Kyoto University’s Institute for the Advanced Study of Human Biology (WPI-ASHBi) has developed a new computational method called Species-OT. By comparing the “shapes” formed by gene expression patterns, it identifies candidate genes that may play similar roles across species and measures overall differences between datasets.

The team tested Species-OT using data on germ cell specification in mice, cynomolgus macaques, and humans, and on pluripotent stem cells from six mammalian species. Among its top predictions were pairs of mouse and human genes that previous studies had shown to play similar roles. These findings can guide further experiments to understand which biological mechanisms are shared across species and which differ.

The findings will be published online in Cell Reports Methods on October 29, 2026.

The behavior of polymer brushes can vary depending on their environment, making their performance difficult to predict. To address this, Tohoku University researchers have proposed a new approach combining polymer physics, machine learning, simulations, and experiments to guide their design.

To prevent the spread of dangerous diseases by mosquitoes, scientists have long studied their reproductive mechanisms to control their population sizes. Now, a team of Nagoya University and University of Iowa researchers have found that the ‘fruitless’ gene in mosquitoes is responsible for the male’s ability to locate females by listening for their wingbeat sound. Mutating this gene in males of the yellow fever mosquito, Aedes aegypti, makes them no longer respond to female wingbeat sounds, disrupting mating behaviors.