BAF60A governs beta cell identity to control systemic glucose homeostasis

By: Shuaishuai Zhu · Ziyin Zhang · Ruo-Ran Wang · Wei Chen · Qiyao Xu · Xinyuan Qiu · Xiuyu Ji · Rui Zhang · Rui Wei · Hongxing Fu · Sheng Yan · Zhuo-Xian Meng

3 October 2026 at 18:00

Diabetologia. 2026 Oct 3. doi: 10.1007/s00125-026-06884-2. Online ahead of print.

ABSTRACT

AIMS/HYPOTHESIS: Chromatin remodelling is critical for maintaining pancreatic beta cell identity and function, yet the key regulatory mechanisms remain incompletely defined. This study aimed to investigate the role of the switch/sucrose non-fermentable (SWI/SNF) complex subunit BAF60A in preserving beta cell fate and glucose homeostasis.

METHODS: Pdx1-Cre-mediated BAF60A-knockout (BaBKO) and BAF60A-overexpressing (BaBOE) mice, together with tamoxifen-inducible adult beta cell-specific Smarcd1 knockout (BaBKO^TM) and Isl1 knockout (Isl1BKO^TM) mice, were generated to evaluate the role of BAF60A in vivo. Glucose homeostasis was assessed through glucose tolerance tests, insulin tolerance tests and glucose-stimulated insulin secretion (GSIS) assays. Multiomic analyses, including RNA-seq, ATAC-seq, Cleavage Under Targets and Tagmentation (CUT&Tag) and single-cell RNA-seq, were performed to characterise chromatin accessibility and transcriptional changes. BAF60A-interacting proteins were identified with biotin identification (BioID) and GST pull-down assays. Beta cell lineage tracing was used to assess changes in cell identity. In addition, BAF60A and the dedifferentiation marker ALDH1A3 were examined in pancreatic islets from individuals with and without type 2 diabetes.

RESULTS: BaBKO mice exhibited significant glucose intolerance, impaired GSIS and pronounced loss of beta cell identity, accompanied by the acquisition of non-beta endocrine features. Inducible deletion of Smarcd1 in adult beta cells similarly impaired beta cell maturation and promoted dedifferentiation, as confirmed by lineage tracing. BAF60A deficiency reduced enhancer accessibility and downregulated beta cell identity genes. Mechanistically, BAF60A physically interacts with the transcription factor islet-1 (ISL1) to regulate transcription of target genes. Adult beta cell-specific Isl1 deletion recapitulated key features of BAF60A deficiency and abolished the beneficial effect of BAF60A overexpression on insulin secretion. Conversely, BaBOE mice exhibited improved glucose tolerance and enhanced GSIS under high-fat diet conditions. Adeno-associated virus-mediated BAF60A overexpression markedly reduced beta cell dedifferentiation in BKS-db/db mice. In human type 2 diabetes islets, BAF60A expression was significantly reduced and inversely correlated with ALDH1A3.

CONCLUSIONS/INTERPRETATION: This work establishes BAF60A-ISL1-dependent chromatin remodelling as a key mechanism that preserves beta cell identity and function under metabolic stress, providing mechanistic insight into beta cell failure in type 2 diabetes.

PMID:42829354 | DOI:10.1007/s00125-026-06884-2

Switchable single-atom catalysts for highly selective C–C coupling in direct methane oxidation

Nature Nanotechnology

By: Xiao Han · Peixin Cui · Geng Wu · Jinyan Cai · Chao Wang · Fanfei Sun · Panzhe Qiao · Xiaocheng Liu · Jiahao Zhuang · Huan Liu · Fabing Su · Gongming Wang · Wensheng Yan · Zeming Qi · Yue Lin · Xun Hong · Yadong Li

7 September 2026 at 08:00

Nature Nanotechnology, Published online: 07 September 2026; doi:10.1038/s41565-026-02271-5

Single copper atoms on boron nanosheets dynamically and reversibly switch to clusters, enabling the direct conversion of methane to acetic acid with 97% selectivity and high activity without the requirement for carbon monoxide.

Mastering Negation: Boosting Grounding Models via Grouped Opposition-Based Learning

cs.AI, q-bio.NC updates on arXiv.org

By: Zesheng Yang · Xi Jiang · Bingzhang Hu · Weili Guan · Runmin Cong · Guo-Jun Qi · Feng Zheng

16 March 2026 at 12:00

arXiv:2603.12606v1 Announce Type: cross Abstract: Current vision-language detection and grounding models predominantly focus on prompts with positive semantics and often struggle to accurately interpret and ground complex expressions containing negative semantics. A key reason for this limitation is the lack of high-quality training data that explicitly captures discriminative negative samples and negation-aware language descriptions. To address this challenge, we introduce D-Negation, a new dataset that provides objects annotated with both positive and negative semantic descriptions. Building upon the observation that negation reasoning frequently appears in natural language, we further propose a grouped opposition-based learning framework that learns negation-aware representations from limited samples. Specifically, our method organizes opposing semantic descriptions from D-Negation into structured groups and formulates two complementary loss functions that encourage the model to reason about negation and semantic qualifiers. We integrate the proposed dataset and learning strategy into a state-of-the-art language-based grounding model. By fine-tuning fewer than 10 percent of the model parameters, our approach achieves improvements of up to 4.4 mAP and 5.7 mAP on positive and negative semantic evaluations, respectively. These results demonstrate that explicitly modeling negation semantics can substantially enhance the robustness and localization accuracy of vision-language grounding models.

From Text to Forecasts: Bridging Modality Gap with Temporal Evolution Semantic Space

cs.AI, q-bio.NC updates on arXiv.org

By: Lehui Li · Yuyao Wang · Jisheng Yan · Wei Zhang · Jinliang Deng · Haoliang Sun · Zhongyi Han · Yongshun Gong

16 March 2026 at 12:00

arXiv:2603.12664v1 Announce Type: cross Abstract: Incorporating textual information into time-series forecasting holds promise for addressing event-driven non-stationarity; however, a fundamental modality gap hinders effective fusion: textual descriptions express temporal impacts implicitly and qualitatively, whereas forecasting models rely on explicit and quantitative signals. Through controlled semi-synthetic experiments, we show that existing methods over-attend to redundant tokens and struggle to reliably translate textual semantics into usable numerical cues. To bridge this gap, we propose TESS, which introduces a Temporal Evolution Semantic Space as an intermediate bottleneck between modalities. This space consists of interpretable, numerically grounded temporal primitives (mean shift, volatility, shape, and lag) extracted from text by an LLM via structured prompting and filtered through confidence-aware gating. Experiments on four real-world datasets demonstrate up to a 29 percent reduction in forecasting error compared to state-of-the-art unimodal and multimodal baselines. The code will be released after acceptance.

Maximizing carrier extraction in hybrid back-contact silicon solar cells

Nature - Issue - nature.com science feeds

By: Zilong Zheng · Xiqi Yang · Jiaxing Wang · Qinghua Zeng · Chaohua Zhang · Hong Zhang · Jiarong Huang · Yuhua Wang · Zeguo Tang · Rongkun Zhou · Hongbo Cai · Xiaofei Xu · Shenghou Zhou · Wanyu Lu · Qian Kang · Xiaoqing Chen · Kun Zheng · Yongzhe Zhang · Zhiyong Wang · Yusheng Yang · Jinyan Zhang · Hui Yan

10 March 2026 at 08:00

Nature, Published online: 10 March 2026; doi:10.1038/s41586-026-10351-8

Maximizing carrier extraction in hybrid back-contact silicon solar cells

Learning to Remember: End-to-End Training of Memory Agents for Long-Context Reasoning

cs.AI, q-bio.NC updates on arXiv.org

By: Kehao Zhang · Shangtong Gui · Sheng Yang · Wei Chen · Yang Feng

24 February 2026 at 13:00

arXiv:2602.18493v1 Announce Type: cross Abstract: Long-context LLMs and Retrieval-Augmented Generation (RAG) systems process information passively, deferring state tracking, contradiction resolution, and evidence aggregation to query time, which becomes brittle under ultra long streams with frequent updates. We propose the Unified Memory Agent (UMA), an end-to-end reinforcement learning framework that unifies memory operations and question answering within a single policy. UMA maintains a dual memory representation: a compact core summary for global context and a structured Memory Bank that supports explicit CRUD (create, update, delete, reorganize) over key value entries, enabling proactive consolidation during streaming. To evaluate long-horizon memory behavior, we introduce Ledger-QA, a diagnostic benchmark for continuous state tracking where answers are latent values derived from accumulated updates rather than lo cal span retrieval. Across 13 datasets spanning Ledger-QA, Test-Time Learning, and Accurate Retrieval, UMA substantially outperforms long-context and RAG baselines on dynamic reasoning and learning tasks while remaining competitive on standard retrieval benchmarks, underscoring the importance of learned, end-to-end memory management.

Reading view