Development of ShotgunCSP: a crystal structure prediction algorithm combining machine learning and first-principles calculations.Achieved world-leading performance in crystal structure prediction benchmarks.A machine learning algorithm for predicting crystal symmetry dramatically improves the performance of structural predictions for complex and large-scale crystal systems.

Developed the machine learning algorithm E2T and its software for learning to learn for extrapolative prediction.Achieved outstanding extrapolative prediction performance in material property prediction tasks across diverse material systems.Demonstrated that models exposed to extensive extrapolative tasks can acquire the ability to rapidly adapt to new tasks.

An Osaka Metropolitan University researcher has developed an autonomous driving algorithm for agricultural robots used for greenhouse cultivation and other farm work.

Recent studies have revealed that electrons passing through chiral molecules exhibit significant spin polarization--a phenomenon known as Chirality-Induced Spin Selectivity (CISS). This effect stems from a nontrivial coupling between electron motion and spin within chiral structures, yet quantifying it remains challenging.

To address this, researchers at the Institute for Molecular Science (IMS) /SOKENDAI investigated an organic superconductor with chiral symmetry. They focused on nonreciprocity related to spin-orbit coupling and observed an exceptionally a large nonreciprocal transport in the superconducting state, far exceeding theoretical predictions. Remarkably, this was found in an organic material with inherently weak spin-orbit coupling, suggesting that chirality significantly enhances charge current-spin coupling with inducing mixed spin-triplet Cooper pairs.

Why does dementia affect more women than men? To help solve this mystery, researchers uncovered a new risk factor: age of menopause onset.

“Do we see colors the same way?” is a fundamentally human question and one of great importance in research into the human mind. While impossible to answer at present, researchers from the University of Tokyo and from Monash University in Australia take steps to answering it using a method that can map the experiences of colors between individuals, including those with colorblindness.

Hepatocellular carcinoma (HCC) is a primary liver cancer that often occurs in patients with non-viral chronic liver disease (CLD). In order to identify specific pathways that are dysregulated in CLD, researchers compared the gene expression and metabolites of CLD liver tissue from patients with HCC to normal, healthy liver tissue. The study demonstrated that inflammatory and fatty acid oxidation pathways are altered in CLD, providing potential therapeutic targets for HCC prevention.

Gene sequencing studies have uncovered mutations in key oncogenes responsible for colorectal cancer (CRC) development. CRC tumors exhibit significant heterogeneity, and a subset of tumors harbor other key genetic and epigenetic alterations that drive carcinogenesis. Now, researchers from Japan have characterized gene mutations in CRC tumors with high tumor mutation burden that lack mutations in the major oncogenes. Their findings highlight alternate site-specific mechanisms of CRC development that can guide treatment selection.

Researchers have developed a machine learning workflow to optimize the output force of photo-actuated organic crystals. Using LASSO regression to identify key molecular substructures and Bayesian optimization for efficient sampling, they achieved a maximum blocking force of 37.0 mN—73 times more efficient than conventional methods. These findings could help develop remote-controlled actuators for medical devices and robotics, supporting applications such as minimally invasive surgery and precision drug delivery.