Core Viewpoint - The article discusses the introduction of Emu3, a multimodal large model developed by Beijing Academy of Artificial Intelligence, which aims to unify the learning of text, images, and videos through next-token prediction, potentially transforming the AI landscape [2][3]. Multimodal Learning - Multimodal learning refers to the ability of AI to process various types of information simultaneously, akin to human sensory perception. Achieving a unified algorithm for learning and generating content from multiple modalities has been a long-standing challenge in the AI field [6]. Emu3's Mechanism - Emu3 employs a simple yet effective approach by converting all modal data into discrete tokens and using a Transformer model to predict the next token, which is a key factor in the success of GPT series language models [6][7]. Training Process - The training of Emu3 consists of three stages: 1. Pre-training with large-scale multimodal data, balancing the loss weights of text and visual tokens to prevent dominance of visual tokens [10]. 2. Post-training for quality fine-tuning on generation tasks, incorporating human preference optimization [10]. 3. Inference supporting classifier-free guidance for low-latency and high-throughput generation [11]. Performance Comparison - Emu3 has demonstrated performance that matches or exceeds specialized models across various tasks: - In image generation, it achieved a human preference score of 70.0, surpassing Stable Diffusion v1.5 (59.3) and SDXL (66.9) [13]. - In video generation, it scored 81.0 in VBench evaluation, comparable to mainstream diffusion models [13]. - In visual language understanding, it averaged 62.1 across 12 benchmark tests, rivaling models like LLaVA-1.6 [13]. - In robotic operations, it achieved a success rate of 87.0% in a simulated environment [13]. Significance of the Research - The significance of Emu3 lies not only in its performance improvements but also in its simplification of paradigms. It demonstrates that next-token prediction can serve as a core paradigm for multimodal models, paving the way for the development of more powerful "world models" that integrate perception, language, and action [15][17]. Future Developments - Following Emu3, the research team has introduced Emu3.5, which enhances the model's capabilities through large-scale long-sequence video training, improving its ability to model physical world dynamics and observing trends in multimodal capabilities as the model and data scale increase [15].

Core Viewpoint - The article discusses the revolutionary AAVLINK technology, which overcomes the limitations of adeno-associated virus (AAV) in gene therapy by enabling the delivery of large genes, thus providing new treatment possibilities for genetic diseases such as autism and epilepsy [2][27]. Group 1: AAVLINK Technology Overview - AAVLINK stands for "AAV with translocation linkage," utilizing the Cre/lox system for DNA recombination, allowing large genes to be split into smaller segments and delivered via multiple AAVs [4]. - The technology enables precise reassembly of these gene segments within cells, leading to the expression of functional proteins [4]. Group 2: Advantages of AAVLINK - AAVLINK demonstrates significant advantages over existing methods for delivering large genes, such as protein splicing and RNA splicing, which often have low efficiency and produce many by-products [6]. - In experiments, AAVLINK achieved over 25 times higher efficiency in reconstructing fluorescent proteins compared to the intein method, and up to 245 times higher in triple vector delivery [12]. - AAVLINK produces minimal by-products, ensuring safety by avoiding truncated proteins that could interfere with normal functions [12]. Group 3: Application in Disease Models - AAVLINK was tested in two disease models: Phelan-McDermid syndrome (PMS) and Dravet syndrome, successfully delivering and reconstructing the SHANK3 and SCN1A genes, respectively, leading to improved behavioral and survival outcomes in mouse models [8][11]. Group 4: CRISPR Delivery Capability - AAVLINK can also deliver large CRISPR-Cas systems, enabling gene editing and regulation, which opens new avenues for treating genetic disorders, such as lowering cholesterol levels by editing the PCSK9 gene [11]. Group 5: Safety Enhancements and Resource Availability - The development of AAVLINK 2.0 addresses potential safety risks associated with the long-term presence of Cre enzyme in the body, paving the way for clinical applications [15][17]. - AAVLINK has established a resource library containing 198 large disease-related genes and 5 CRISPR tools, allowing researchers to accelerate therapy development [19][21].

Core Viewpoint - The research highlights a novel therapeutic approach targeting KRAS-driven acute myeloid leukemia (AML) by selectively depleting mitochondrial NAD+ through dual inhibition of SLC25A51 and succinate dehydrogenase (SDH) [2][6]. Group 1: Research Findings - The study identifies that KRAS mutations drive high-proliferation AML cells that are resistant to existing treatments and glucose stress [3]. - Compound-615 was found to selectively eliminate KRAS mutant cells by simultaneously inhibiting SDH and the mitochondrial NAD+ transporter SLC25A51, leading to mitochondrial NAD+ depletion [3][8]. - Mechanistically, KRAS mutant cells exhibit reduced succinylation at the K264 site of SLC25A51, which is crucial for protein stability, creating a synthetic lethal vulnerability [3][8]. Group 2: Implications of the Study - The research reveals a specific metabolic vulnerability in KRAS-driven AML, suggesting a dual inhibition therapy as a promising treatment strategy [6]. - KRAS wild-type cells maintain sufficient NAD+ influx due to high expression of SLC25A51, allowing them to evade the cytotoxic effects of Compound-615 [8].

编辑丨王多鱼 排版丨水成文 糖尿病的细胞治疗 长期受限于一个 悬而未决 的根本难题—— 移植后的 胰岛 β 细胞难以在糖尿病病理 微 环境中实现长期存活并维持稳定功能 。临床随访数据表 明，多数胰岛移植在约 5 年后出现显著的 β 细胞功能衰退与治疗效果减弱。 干细胞来源的 β 细胞 （SC- β细胞 ） 也面临相同的困境。 这一移植后的" 长期 存活瓶颈"，已成为制约糖尿病细胞治疗进一步走向临床应用与规模化推广的关 键因素。如何突破该瓶颈，实现具备长期稳定疗效、甚至"一次性"治愈潜力的治疗方案，正是当前糖尿病细胞治疗领域亟需解决的核心目标。 该研究揭示了在糖尿病发生发展过程中， 锌离子异常积累 是导致 β 细胞身份丢失的核心驱动因素 ， 并 证实介导 锌离子积累 的转运蛋白 Z nT8 是 糖尿病预防 与治疗 的重要靶点。该研究发现， β 细胞内 过载的锌 持续 激活整合应激反应 （ ISR） ， 并 通过 ISR 下游转录因子 ATF4 启动 α 细胞特异性转录因子 ARX 异位 表达，从而在分子层面推动 β 细胞转分化。 这一发现为糖尿病细胞治疗长期疗效不足提供了全新解释与潜在干预靶点。 基于上述机制， ...

撰文丨王聪 编辑丨王多鱼 排版丨水成文 在对抗癌症的战争中，我们身体的免疫系统本是强大的盟友。化疗和免疫疗法等治疗手段，一个重要目的 就是重新唤醒并武装这个盟友，去攻击肿瘤细胞。然而，现实往往很残酷：许多患者最初对治疗有效，但 随着时间推移，肿瘤会产生耐药性，导致治疗失败、肿瘤复发。 这背后，是肿瘤细胞在不断地学习和进化，从而"伪装"自己，逃逸免疫系统的追杀。然而，其中的 潜在机 制，在很大程度上仍不为人知。 2026 年 1 月 26 日， 武汉大学 蓝柯 教授团队 （ 蓝柯教授 、 邬开朗副教授 、 祝成亮 主任技师 为论文 共同通讯作者 ； 武汉大学田明富博士、刘思雨博士研究生、李旭博士为论文共同第一作者 ） 在 Vita 期刊 发表了题为： PNPT1-mtRNA axis mediates chemotherapy-induced immune signaling and can be targeted to overcome therapeutic resistance 的研究论文。 该研究系统揭示了 线粒体 RNA （mtRNA） 衍生的 危险信号 在激活 天然 免疫 和 抗肿瘤治疗中的关键作 ...

编辑丨王多鱼 排版丨水成文 铁死亡 （ Ferroptosis ） 是 一种由铁依赖性脂质过氧化驱动的细胞程序性死亡形式， 在 过去十余年迅速 成为生物医学研究的热点。在肿瘤、神经退行性疾病和急性肾损伤等多种疾病中，铁死亡正逐渐成为具有 临床应用前景的调控通路。 然而，长期以来一个被忽视的问题是：活性氧 （ ROS） 诱导剂是否必然推动铁死亡？进一步而言，在评 估铁死亡抑制剂时，除其还原性外，是否还需要系统检验其氧化性属性？ 2026 年 1 月 27 日，暨南大学生命科学技术学院 黄俊祺 、浙江大学爱丁堡大学联合学院 Chew Ting Gang 及湛江中心人民医院 庞丽娟 、 郭允苗 作为共同通讯作者 （ 乔梦浩 、 周丽群 为论文共同第一作者 ） 在 Advanced Science 期刊发表了题为： ERM Inhibition Confers Ferroptosis Resistance through ROS-Induced NRF2 Signaling 的研究论文。 该研究揭示， ERM （ E zrin- R adixin- M oesin ） 家族蛋白是铁死亡敏感性的调控开关， ERM- 肌 ...

撰文丨王聪 编辑丨王多鱼 排版丨水成文 人们越来越意识到， 肠神经系统 （ Enteric Nervous System， ENS ） 可能是 帕金森病 （PD） 中 α- 突触核蛋白 （ α-Synuclein， αS ） 病理变化的最初发生部位。 便秘 是帕金森病患者最早出现且最常见 的症状之一，这可能反映了肠神经系统功能障碍。 死后解剖和多模态影像学研究显示，帕金森病患者的肠神经系统存在早期 路易体病理 （ Lewy Pathology ） ，其从肠道向脑干及其他脑区呈尾向头的分布，这一发现有力地支持了帕金森病的" 体优先 " （ body- first ） 观点。此外，在小鼠中进行肠道注射 α-突触核蛋白 会导致路易体病理现象按层级向大脑 扩散，这 有力地表明肠神经系统在帕金森病早期阶段就已受到牵连。 然而，对于可能引发肠神经系统中 α-突触核蛋白病理变化的起始以及其向大脑发展的细胞和分子机制，目 前仍知之甚少。 2026 年 1 月 28 日， 伦敦大学学院的研究人员在国际顶尖学术期刊 Nature 上发表了题为： Intestinal macrophages modulate synucleino ...

Core Viewpoint - The research team led by Lei Xiaoguang from Peking University has made a significant breakthrough in the field of amide bond synthesis by engineering aldehyde dehydrogenases to create a new biocatalytic strategy for amide bond formation, which is more efficient and environmentally friendly compared to traditional methods [2][4][11]. Group 1: Research Breakthrough - The team successfully transformed aldehyde dehydrogenase (ALDH) into an oxidative amidase (OxiAm) that catalyzes the direct reaction of aldehydes with amines to form amides, eliminating the need for traditional coupling agents that produce waste [4][5]. - This new biocatalytic method allows for the construction of amide bonds from low oxidation state precursors without generating harmful waste, promoting a sustainable approach to drug manufacturing [4][11]. Group 2: Mechanism and Innovation - The research fundamentally restructured the classical ALDH catalytic pathway by modifying the enzyme's active site to favor amine substrates over water, leading to direct amide formation [5][8]. - The team introduced a two-step enzyme cascade reaction system that first oxidizes alcohols to aldehydes, which are then converted to amides, significantly broadening the range of starting materials [8][10]. Group 3: Implications for Drug Development - The new strategy was applied to redesign the synthesis route of Imatinib (Gleevec), a key drug for leukemia treatment, demonstrating advantages in reducing reaction steps, minimizing byproduct formation, and enhancing overall atom economy [8][10]. - The innovative approach represents a shift towards greener and more efficient methods in pharmaceutical synthesis, aligning with the growing emphasis on sustainable practices in the chemical industry [18].

撰文丨王聪 编辑丨王多鱼 排版丨水成文 在屏障组织中， 免疫监视 对机体维持稳态、抵御感染并清除早期异常细胞尤其重要。CD8 + 组织驻留记忆 T 细胞 （T RM ） 在屏障组织中长期驻留，能够在第 一时间识别并应对感染或肿瘤发生，是决定组织局部免疫监视强度的关键细胞群。 长期以来， 压力状态常伴随免疫波动 ，但系统性的压力信号如何被传递到组织器官内部，并影响局部免疫监视的强弱，仍缺乏明确的细胞与分子机制。 2026年1月29日，西湖大学生命科学学院 张兵 团队与西湖大学医学院 周挺 团队合作 （ 博士生 张鹏 、 缪菊菊 为论文共同第一作者 ） ，在国际顶尖学术期刊 Cell 上发表了题为： Sympathetic-Epithelial Crosstalk Governs Tissue-Resident Memory T Cell Immunosurveillance in the Skin 的研究论文。 该研究首次发现， 交感神经 能够通过"遥控"角质形成细胞，间接调控皮肤中的 CD8 + 驻留记忆 T 细胞 （T RM ） 丰度，从而影响皮肤对癌症的免疫监视能力。 研究团队进一步证实，在小鼠黑色素瘤模 ...

Core Insights - On January 28, 2026, a total of 43 research papers were published in the prestigious journal Nature, with 22 of them authored by Chinese scholars, highlighting the significant contribution of Chinese researchers to global scientific advancements [3][4][6][8][10][11][12][14][18][19][22]. Group 1: Research Contributions - The paper titled "Constraints on axion dark matter by distributed intercity quantum sensors" was authored by Professor Peng Xinhua and Professor Jiang Min from the University of Science and Technology of China, with postdoctoral researcher Wang Yuanhong as the first author [3]. - A study on "Prethermalization by random multipolar driving on a 78-qubit processor" was published by researchers from the Chinese Academy of Sciences and Peking University, with Liu Zhenghe and Liu Yu as co-first authors [4]. - The research "Multimodal learning with next-token prediction for large multimodal models" was led by Professor Huang Tiejun and Wang Zhongyuan from the Beijing Academy of Artificial Intelligence, with Wang Xinlong as a co-first author [6]. - The paper "Radiation-tolerant atomic-layer-scale RF system for spaceborne communication" was authored by Professor Zhou Peng and Associate Professor Ma Shunli from Fudan University, with postdoctoral researcher Zhu Liyuan as the first author [8]. - The study "Accurate determination of the 3D atomic structure of amorphous materials" was published by Miao Jianwei from UCLA, with Liao Yuxuan as the first author [10]. Group 2: Diverse Research Topics - The research titled "Optical control of integer and fractional Chern insulators" was authored by Xu Xiaodong from the University of Washington, with Li Weijie as a co-first author [11]. - The paper "Bandwidth-tuned Mott transition and superconductivity in moiré WSe2" was co-authored by researchers from Cornell University, with Xia Yiyu and Han Zhongdong as co-first authors [12]. - The study "Frequency reproducibility of solid-state thorium-229 nuclear clocks" was published by Ye Jun from the University of Colorado Boulder [13]. - The research "A Cambrian soft-bodied biota after the first Phanerozoic mass extinction" was authored by Zhu Maoyan and Zhao Fangchen from the Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences [14]. - The paper "Advancing regulatory variant effect prediction with AlphaGenome" was co-authored by Cheng Jun from Google DeepMind, with co-first author Zhang Mingchao [16].