A computational method called scSurv, developed by researchers at Institute of Science Tokyo, links individual cells to patient outcomes using widely available bulk RNA sequencing data. The approach uses single-cell reference datasets together with patient survival data to infer the contributions of individual cells within complex tissues. The model identified cell populations associated with survival across several cancers, offering a way to uncover disease-driving cells and support the development of more targeted treatment strategies.

Kyoto, Japan -- Our genes are written in long strings of three-letter units composed of four different nucleotides. These units -- or codons -- specify one of many amino acids, the building blocks of proteins. Multiple codons can encode the same amino acid, which seems to point to some redundancy in our genetic code.

Yet growing evidence suggests that these synonymous codons are not interchangeable: rather, some confer stability to mRNAs and are more efficiently translated in cells, and thus more optimal than others. mRNAs enriched in non-optimal codons are inefficiently translated and subsequently degraded, but how human cells detect and respond to these substandard codons has largely remained a mystery.

A collaborative team of researchers at Kyoto University and RIKEN, led by Osamu Takeuchi and Takuhiro Ito, was determined to unravel this enigma, and conducted several tests to better understand this process.

A multinational research team led by researchers at Institute of Science Tokyo, RIKEN, and the University of Toronto has revealed how a tryptophan-rich allosteric communication network regulates receptor dynamics and activation of the human adenosine A_2A receptor (A_2AR), a major G protein-coupled receptor (GPCR) drug target. By integrating experimental functional assays and residue-specific NMR with molecular simulations and fast allostery-prediction algorithms based on rigidity theory, the team mapped long-range allosteric communication pathways linking the ligand-binding pocket to the intracellular G protein–coupling machinery and identified a central role for tryptophan residues along these pathways. The study also clarifies the functional role of the receptor’s conserved sodium-binding pocket, showing that sodium egress strongly promotes activation-related conformational states, including a precoupled state that likely prepares the receptor for productive G protein interaction. These findings deepen our understanding of GPCR activation and allostery, and may support future development of allosteric GPCR drugs.

Beyond the specific mechanism, this work addresses a major bottleneck for AI in structural biology: recent advances such as AlphaFold have transformed prediction of static protein structures, but AI still cannot reliably predict the dynamics and allosteric communication that determine function, signaling, and drug response. To help close this gap, the researchers developed and applied fast computational methods for probing allosteric and dynamic regulation in protein structures and anchored these predictions with experimental NMR validation. The resulting experimentally validated, computationally generated data on allostery and dynamics—and a scalable approach to extend these datasets across diverse receptors and conditions—provide scarce, high-value training and benchmarking data for next-generation AI models aimed at predicting protein function beyond static structure, accelerating future AI-driven prediction of protein function and the design of selective GPCR therapeutics.

Regulations for rotational atherectomy (RA) were relaxed for Japanese hospitals in 2020, allowing smaller hospitals to perform RA during percutaneous coronary intervention (PCI) for calcified coronary artery disease. Researchers analyzed nationwide registry data and found adverse event rates among patients receiving PCI did not increase in smaller hospitals newly accredited for RA use after 2020 revision, although more patients received RAs. These findings support the safety of the policy change, improving access to advanced treatments.

Second-language sentence processing raises fundamental questions about whether learners rely on native-like structural mechanisms or alternative strategies. In a new study, researchers compared native English speakers and native Japanese speakers learning English as a second language using eye-tracking during comprehension of ambiguous filler-gap sentences. Results revealed that learners with higher structural computation ability showed native-like, structure-based prediction, whereas lower-accuracy learners relied more on lexical cues, supporting a gradient view of predictive processing.

Researchers show that synaptic wiring patterns alone can identify neuron types in fruit fly brain connectomes. Their method, NTAC, assigns neuronal types based solely on synaptic connectivity. With just 2% of neurons pre-labeled, it exceeds 90% accuracy in the fruit fly visual system and runs in minutes on a standard computer. A fully label-free version still reaches about 70% accuracy, enabling scalable cell typing as connectome datasets grow.

Researchers developed high-performance, lead-free piezoelectric thin films on silicon that convert everyday vibrations into electricity. By using strain engineering and a novel sputtering method, they achieved record performance and demonstrated the compatibility of efficient, battery-free MEMS energy harvesters with standard semiconductor manufacturing processes. 

Researchers in Japan have analyzed 10 years (2014-2023) of game data from Japan’s Pacific League and found that the presence or absence of the designated hitter (DH) system does not change the relationship between player talent and winning. Published in PLOS One, the study also presents a new method to measure player contributions by using performance data to compare starting players with backup players at each position.

Utilization of biomass as a chemical resource is a promising strategy for establishing a circular economy. Cellulose, a polymer composed of glucose units, is the most abundant form of biomass, and glucose is a versatile feedstock for chemicals. However, cellulose is a highly recalcitrant material due to its extensive hydrogen-bond (H-bond) network. Kobayashi, Nishimura, and their colleagues discovered that simply dipping cellulose in an aqueous NaOH solution below −28 ℃ makes it more reactive. Thereby, the efficiency of saccharification (the process of breaking down cellulose into sugars) increased 2.2 times.