在生命科学的浩瀚星海中， 空间转录组学 正像一束穿透迷雾的光，把"细胞在哪里、在做什 么、在与谁对话"从想象变成可测量的事实。它让我们第一次有机会把分子表达、组织结构与 细胞命运放在同一张坐标系里理解——而这也意味着，真正的挑战不再是"有没有数据"，而 是"能不能把数据变成机制、把机制变成主图"。 我们相信，未来的空间组学竞争不在"知道使用哪个软件"，而在于谁能更快构建可解释的空间 坐标系，谁能更稳建立"位置-互作-信号-表型"的证据链，谁能把结果做成审稿人一眼认可的 主图。愿这门课程成为你2026开年的第一份"科研加速器"：让复现不再靠运气，让机制不再 停留在猜想，让每一次分析都能沉淀为可复用、可扩展、可发表的能力与作品。 论文介绍 本次复现的论文如下 4.源码级系统学习： https://www.nature.com/articles/s41586-024-08466-x 01 课程特色 1.全流程系统教学： 从空间转录组数据结构、QC、注释到出图全流程，配套"保姆级"R/Python代码模板，快速补齐空间数据分析 底层能力 2.一对一指导 + 包教包会： 针对你自己的数据（Xenium/VisiumHD ...

编辑丨王多鱼 排版丨水成文 细胞的功能不仅取决于其自身的基因表达，更取决于其在三维 （3D） 空间中的位置以及与周围微环境的 3D 互作，理解组织结构的 3D 复杂性是解析生物功能的关键。然而，当前主流的空间转录组学 （Spatial Transcriptomics，ST） 技术大多局限于二维 （2D） 平面，这种观测维度的局限导致我们难以还原器官 内部基因表达的连续梯度、细胞细胞微环境的立体分布以及精细的细胞互作网络。 尽管研究者们试图在Z轴上堆叠多张 2D 切片以近似三维结构，但受限于高昂的实验成本与有限的实验通 量，不得不对 Z 轴方向的采样密度做出妥协。这种折衷方案导致切片之间往往存在不可忽视的物理间距 （例如100微米 ，相当于缺失了约 5-10 层细胞的信息 ） ，使得最终获取的数据在Z轴上呈现为高度稀疏 的采样状态。这种采样方式将原本连续的生物组织割裂为离散的稀疏的断层面，使得现有的三维图谱往往 缺乏连续性与完整性。 2025 年 12 月 31 日，复旦大学 冯建峰 / 原致远 团队联合中国科学院计算技术研究所 赵屹 团队，在 Nature Methods 期刊发表了题为： Bridging ...

Core Viewpoint - The article presents a comprehensive training program focused on bioinformatics, particularly in single-cell analysis and spatial transcriptomics, aimed at enhancing research capabilities in the life sciences [56][59]. Group 1: Course Structure - The training consists of multiple sections covering Python programming, data structures, and advanced analysis techniques using various tools like scanpy and Seurat [2][6][32]. - Specific topics include spatial transcriptomics applications, data normalization, clustering, and trajectory analysis [8][12][18]. Group 2: Practical Applications - The program emphasizes hands-on experience, allowing participants to analyze real datasets and replicate findings from published research [24][25][32]. - It includes practical sessions where students can apply learned techniques to their own data, ensuring a thorough understanding of the analysis process [24][32]. Group 3: Instructor Expertise - The instructor, with extensive experience in medical artificial intelligence and bioinformatics, has guided numerous students in publishing high-impact research articles [56][59]. - The program is designed to provide personalized support, ensuring that all participants can effectively learn and apply the concepts taught [59][70]. Group 4: Community and Support - The training fosters a collaborative environment, encouraging participants to engage with each other and the instructor for ongoing support [70][72]. - There is a commitment to continuous learning, with resources available even after the course concludes, allowing for long-term skill development [72][74].

Core Insights - The collaboration between the University of California, San Francisco, and the Allen Institute has led to the development of an AI model named CellTransformer, which has created the most detailed mouse brain map to date, encompassing 1,300 brain regions and subregions [1][3] Group 1: AI Model and Technology - CellTransformer utilizes a Transformer architecture similar to that used in models like ChatGPT, which excels in understanding contextual relationships [3] - The model analyzes the relationships between adjacent cells in spatial contexts, predicting molecular characteristics based on a cell's "neighborhood structure" [3] Group 2: Brain Mapping Advancements - Unlike previous brain maps that primarily categorized based on cell types, this new model focuses on the brain's structural regions, automatically defining boundaries based on cellular and molecular features rather than human judgment [3][4] - The resulting brain map is one of the most precise and complex data-driven maps of an animal brain to date, accurately representing known regions like the hippocampus and discovering new subregions in less understood areas like the midbrain reticular formation [3][4] Group 3: Implications and Applications - The new brain region delineation is entirely data-driven, revealing numerous unknown areas that may correspond to unexplored brain functions [4] - The potential applications of the CellTransformer model extend beyond neuroscience, with the algorithm being applicable to other organ systems and cancer tissues, utilizing spatial transcriptomics data to uncover biological mechanisms in health and disease, thus providing new tools for drug development and disease treatment [4]

Core Insights - The article discusses a significant research study on esophageal squamous cell carcinoma (ESCC), highlighting the discovery of a GPR116⁺ pericyte subpopulation that plays an immunosuppressive role in cancer metastasis [2][3][5]. Group 1: Research Findings - The study utilized single-cell multi-omics and spatial transcriptomics techniques, analyzing 16 samples with 117,169 cells and 5 samples with 195,366 cells, respectively, to reveal the cellular and spatial characteristics of ESCC [5]. - A pericyte subpopulation driven by the transcription factor PRRX1 was identified, which promotes tumor metastasis and immune evasion [3][5][6]. - GPR116⁺ pericytes secrete EGFL6, which binds to integrin β1 on cancer cells, activating the NF-κB signaling pathway and facilitating tumor progression [6]. Group 2: Clinical Implications - The study suggests that serum levels of EGFL6 could serve as a non-invasive biomarker for the diagnosis and prognosis of various tumors [6]. - Blocking integrin β1 in ESCC animal models effectively inhibited tumor metastasis and significantly improved responses to immunotherapy [6]. - The research provides a spatially resolved map of the tumor microenvironment in ESCC, revealing the biological and clinical significance of GPR116⁺ pericytes, which may lead to innovative treatment strategies for metastatic cancer [6][7].

Core Insights - The collaboration between the University of California, San Francisco, and the Allen Institute has led to the development of an AI model named CellTransformer, which has created the most detailed mouse brain map to date, encompassing 1,300 brain regions and subregions [1] - This new model utilizes a Transformer architecture similar to that used in large models like ChatGPT, allowing for the analysis of spatial relationships between adjacent cells to predict molecular characteristics and construct a detailed brain organization atlas [1] - Unlike previous brain maps that were primarily based on cell types, this new approach focuses on the brain's structural organization, automatically defining boundaries based on cellular and molecular features rather than relying on human judgment [1] Summary by Sections AI Model Development - CellTransformer accurately reproduces known brain regions such as the hippocampus and identifies new, more subdivided subregions in less understood areas like the midbrain reticular formation [2] - The model's data-driven approach reveals numerous unknown regions, which are likely associated with unexplored brain functions, akin to transforming a map that only shows continents and countries into one that includes cities [2] Broader Applications - The potential of the CellTransformer model extends beyond neuroscience, with its algorithms being applicable to other organ systems and even cancer tissues, utilizing spatial transcriptomics data to uncover biological mechanisms in health and disease [2]

Core Insights - The article discusses the global obesity crisis, highlighting that over 1 billion people are affected by obesity, which is linked to various metabolic diseases and health risks [6] - It emphasizes the importance of understanding the changes in adipose tissue during weight loss and the underlying mechanisms that contribute to improved health outcomes [6][18] Group 1: Adipose Tissue Remodeling - A comprehensive single-cell atlas of adipose tissue was created, analyzing over 171,000 cells from 49 individuals, revealing significant immune cell infiltration and a decrease in mature adipocytes in obese individuals [8] - Weight loss was shown to alleviate pathological changes in adipose tissue, indicating a potential for recovery [8] Group 2: Macrophage Memory Effect - The proportion of macrophages in adipose tissue increased from 14% to 31% in obesity, with a notable presence of lipid-associated macrophages (LAMs) [10] - Following weight loss, macrophage numbers decreased to 18%, but their metabolic activation state did not fully revert, suggesting an "epigenetic memory" [10] Group 3: Metabolic Reboot of Adipocytes - Eight subtypes of mature adipocytes were identified, with stress and fibrotic types increasing in obesity, while lipid-synthesis types decreased [12] - Weight loss significantly reduced the stress-type adipocytes and restored lipid-synthesis types, enhancing overall metabolic flux and potentially improving insulin sensitivity [12] Group 4: Decline of Stress Ecological Niche - The proportion of stress-type adipocyte precursor cells (APCs) increased in obesity, while weight loss significantly reduced these cells and downregulated hypoxia-related signals [14] - The vascular system also showed signs of stress, with weight loss reducing the proportion of stressed vascular cells [14] Group 5: Reversal of Aging Programs - Weight loss led to a significant reduction in aging markers across various cell types, indicating a reversal of aging processes [17] - The study identified a conserved transcription factor network in stressed aging cells that could be "turned off" through weight loss, promoting cellular health [17] Group 6: Implications for Future Research - This groundbreaking research not only elucidates the molecular mechanisms by which weight loss improves health but also opens new avenues for developing drug interventions that mimic weight loss effects [18]

Core Insights - The article highlights the significant contributions of Chinese scholars in top international academic journals, with 8 out of 18 papers published in Nature on October 1, 2025, authored by Chinese researchers [2][5][6][7][8][10][12][14]. - A notable paper from Yale University discusses a new method in spatial transcriptomics, RAEFISH, which achieves whole-genome coverage and single-molecule resolution, marking a significant advancement in the field [16][19]. - A study from the National Laboratory of Yacheng Bay reveals the genetic selection trajectories in soybean domestication, providing new insights into breeding strategies and genetic resources [20][23][24]. Group 1: Contributions to Nature - On October 1, 2025, multiple papers authored by Chinese scholars were published in Nature, including significant studies on T cell exhaustion, dietary impacts on intestinal stemness, and new paradigms in protein biogenesis [2][5][6][7][8][10]. - The research from Ohio State University on T cell exhaustion highlights the role of proteotoxic stress in immune evasion [2]. - The study from MIT explores how dietary cysteine enhances intestinal stemness through CD8 T cell-derived IL-22 [5]. Group 2: Innovations in Spatial Transcriptomics - The Yale University team developed RAEFISH, a new spatial transcriptomics method that allows for whole-genome coverage and single-molecule resolution, addressing previous limitations in the field [16][19]. - This advancement provides a powerful tool for various biological research areas, including developmental biology and drug discovery [19]. Group 3: Soybean Genetic Research - The research team at Yacheng Bay National Laboratory studied 8,105 soybean accessions, revealing key genetic selections during domestication and improvement processes [20][23]. - The findings indicate the existence of two independent centers of soybean domestication and highlight the importance of black soybean in this process [23]. - The study also provides insights into the changing breeding priorities in China, emphasizing high protein content in the early years and more recently focusing on yield, oil content, and stress resistance [23][24].

Core Insights - The research conducted by the Salk Institute has established the first comprehensive genetic map of Arabidopsis thaliana, covering its entire life cycle, which will significantly advance plant biology and biotechnology [2][5] Group 1: Research Significance - Arabidopsis thaliana has been a model organism in plant biology for decades, providing insights into plant responses to light, hormonal control, and root structure [2][3] - The new genetic map includes gene expression patterns from over 400,000 cells throughout the plant's life cycle, offering unprecedented information for future studies on plant cell types and developmental stages [2][5] Group 2: Technological Advancements - The integration of single-cell RNA sequencing with spatial transcriptomics allows for the preservation of the plant's original tissue structure, enabling precise localization of gene expression without disrupting cellular arrangements [4][5] - This advanced methodology has led to the creation of a complete gene expression map across ten key developmental stages of Arabidopsis, revealing the remarkable diversity of cell types and the dynamic evolution of gene regulatory networks [5][6] Group 3: Future Implications - The research is expected to serve as a powerful tool for new discoveries in plant biology, with a user-friendly web application developed for global access to the life cycle map [6] - The findings aim to deepen the understanding of plant developmental mechanisms and assist in exploring how plants respond to genetic variations and environmental pressures, thereby promoting advancements in crop improvement and ecological adaptation research [6]

Core Insights - The article emphasizes the importance of understanding obesity through advanced scientific techniques, particularly single-nucleus RNA sequencing and spatial transcriptomics, which provide a detailed view of cellular changes in adipose tissue [6][7][12] Group 1: Research Findings - The study included three groups: 24 healthy individuals, and 25 obese individuals before and after weight loss surgery, revealing that weight loss surgery reduced the average BMI from 45.2 to 35.2, significantly improving fasting insulin and insulin resistance [7] - Analysis of over 170,000 cells identified more than 20 different cell states, showing a clear distinction in cellular organization between healthy and obese individuals, with a notable increase in macrophages in obese tissue [7][8] - In obese individuals, macrophages constituted 31% of adipose tissue, compared to 14% in healthy individuals, indicating a shift in immune cell dynamics [8] Group 2: Cellular Dynamics - The study identified two subtypes of lipid-associated macrophages (LAMs) in obese tissue: adaptive LAMs, which efficiently process lipids, and inflammatory LAMs, which are associated with insulin resistance [8][9] - The proportion of "stress-type" adipocytes in obese tissue was found to be 55%, which dropped to 14% post-weight loss, indicating a significant reduction in unhealthy adipocyte types [9][10] - The research linked obesity to cellular senescence, revealing that "stress-type" adipocytes express high levels of the senescence marker p21, which were largely eliminated after weight loss [10] Group 3: Implications for Treatment - The findings suggest that weight loss is not only about reducing fat but also involves a systemic cleansing of senescent cells, enhancing overall tissue health [12] - The persistence of inflammatory macrophages post-weight loss raises concerns about potential metabolic rebound, highlighting the need for preventive strategies [12] - The research provides insights into potential future treatments for obesity, focusing on targeting dysfunctional cells and signaling pathways rather than solely addressing energy balance [12]