Global Intelligence Database
Researchers have developed DReKGNN, a novel framework that enhances drug repositioning by integrating expert biological knowledge into AI models. By processing verified database information through Large Language Models, the system predicts new uses for existing medications more effectively than traditional methods.
By Wu, Yan, Chen, Xiao, Gao
A major new analysis of over 3,000 patients reveals a powerful link between a common blood enzyme and survival rates for a type of liver cancer. Elevated levels of gamma-glutamyltransferase (GGT) before treatment strongly predict poorer outcomes, suggesting it could be a valuable prognostic marker for doctors.
By Huang, Zhou, Feng, Xiao, Zhong
A new software programme called BugBox uses artificial intelligence to rapidly classify arthropods from photographs, scaling up biodiversity research. Although slightly less accurate than human experts, the system learns from corrections and successfully identified that regenerative farming boosts insect diversity.
By Welch, Wilson, Lundgren
Standard silicon electronics degrade rapidly in harsh radiation environments like deep space or nuclear facilities. A new review highlights how carbon nanotubes and magnetic memory offer superior durability due to their unique atomic structures, promising a leap forward for robust computing.
By Zhou, Wang, Zhang, Lin, Zhao