Artificial intelligence is rapidly changing how biomedical materials are discovered and designed. In a new review published in Science Bulletin, researchers summarize how machine learning and generative AI are helping scientists predict material behaviors such as drug release, biocompatibility, and catalytic activity, and even propose entirely new inorganic materials for biomedical applications. The review highlights emerging applications in drug delivery, cancer therapy, anti-inflammatory treatment, and tissue engineering, while also discussing the challenges that still limit clinical translation.

A research team led by Professor Gao Zexia from Huazhong Agricultural University in China has successfully cultivated grass carp completely lacking intermuscular bones (IBs)—small, sharp skeletal structures that not only pose consumption risks to consumers but also limit processing efficiency. By runx2b gene editing, the team obtained genetically stable F₂ populations of grass carp without IBs. Comprehensive phenotypic analysis revealed no significant differences between IBs-free and the wild-type grass carp in overall skeletal development, bone mineral density, muscle composition or nutritional content. Multi-omics data further demonstrated adaptive remodeling in Calcium Signaling and Muscle Contraction Pathways, indicating normal physiological compensation. This study provides a feasible strategy for improving the fish efficiency of aquaculture breeding and enhancing the supply of high quality fish protein.

Researchers from Capital Normal University and Peking University have revealed how light triggers seed germination at the molecular level. This research demonstrates that the HECT E3 ligase UPL3 targets the core repressor PIF1 for degradation upon light exposure, offering theoretical support and promising genetic targets for improving crop seed vigor and germination uniformity.

A recent comprehensive review highlights that artificial intelligence, by integrating multiscale data from wearables, multiomics, imaging and largescale cohorts, can finally turn the “exercise is medicine” slogan into truly individualized clinical practice. This openaccess article entitled “Artificial intelligenceempowered multiscale data integration for precision exercise biomedicine” (DOI: 10.1515/teb20260009), was published in Translational Exercise Biomedicine (ISSN: 2942-6812), an official partner journal of International Federation of Sports Medicine (FIMS).