All Items
-
Abstract:
Iron (Fe) redox cycling is intricately linked to methane (CH₄) emissions in global wetlands, yet its role under sustained bioturbation remains poorly quantified. We investigated how continuous loach (Misgurnus anguillicaudatus) activity influences CH₄ emissions and Fe dynamics in a ratoon rice system over 178 days. Methane and ecosystem CO₂ fluxes were measured continuously, while in situ...
-
Abstract:
The ALKA channel, a member of the Cys-loop ligand-gated ion channel (cysLGIC) family, has recently been identified as a receptor mediating alkaline taste sensation in Drosophila. However, the structure and molecular mechanism underlying pH sensing of ALKA channel remain to be determined. Here, we report the cryo-EM structure of ALKA, determined at a resolution of 2.86 Å under its physiological pH...
-
A continuous cost–efficiency spectrum governs brain-wide communication and cognitive reconfiguration
Abstract:Information communication in the brain must balance efficient functional integration with the metabolic and physical costs of long-range connectivity. How this cost–efficiency balance is implemented in large-scale communication networks, and how it adapts across cognitive states and pathology, remains poorly understood. In this study, by systematically tuning the strength of distance constraints...
-
Abstract:
The nuclear envelope (NE) plays a crucial role in genome organization by tethering heterochromatin to the nuclear periphery. Yet, how individual NE-associating factors regulate 3D genome architecture remains incompletely understood. Here, we leverage the mitosis-to-G1 phase transition as an experimental system to dissect the roles of Lamin A/C and Lamin B receptor (LBR) in post-mitotic genome...
-
Abstract:
Protein evolution in nature and in the laboratory proceeds through incremental, largely undirected mutational steps, restricting exploration to local regions of sequence space and limiting access to remote yet potentially functional proteins. We present EvoGUD, a single-sequence–conditioned diffusion framework for large-step exploration of protein sequence space under learned evolutionary...
-
Abstract:
Both natural and directed evolution are powerful in improving protein functions but they are slow in exploring the nearly endless sequence space. Here, we present SPIN-dvEvo that couples few-shot low-rank adaptation (LoRA) of an ESM-2 protein language model with a genetic algorithm to quickly evolve functional remote homologs from a local cluster of highly-homologous, binary-labeled sequences. We...
-
Abstract:
Proteolysis Targeting Chimera (PROTACs) are a revolutionary drug modality that can expand the repertoire of druggable targets to proteins lacking conventional active sites or binding pockets1. However, the majority of PROTACs developed are against already drugged targets2 suggesting it is challenging to discover novel binders of undruggable targets. To overcome this, we developed a scalable,...
-
Abstract:
Tumor metabolic rewiring supports malignant growth and can concurrently dampen antitumor immunity, yet how the tumor-intrinsic mevalonate (MVA) pathway enforces immune escape remains insufficiently defined. Here we show that genetic or pharmacologic disruption of the MVA pathway in tumor cells elicits a robust immune response, increasing intratumoral CD8⁺ T cell infiltration and effector function...
-
Abstract:
The NLRP3 inflammasome is a key driver in inflammatory, infectious, metabolic, and neurodegenerative diseases. Although the NLRP3 inhibitor CRID3 (also known as MCC950) exhibits potent activity, it cannot inhibit several hyperactive NLRP3 mutations associated with autoinflammatory syndromes and has not progressed clinically, underscoring the need for the development of new NLRP3 inhibitors....
-
Abstract:
Large language models predominantly rely on the Transformer architecture, whose self-attention mechanism incurs a quadratic computational cost O(N2) with respect to input length, leading to significant memory and computation bottlenecks when processing ultra-long contexts. This work proposes LanguageFold, a hierarchical sparse attention mechanism inspired by the Self-Returning Random Walk model...