然而，现有的大规模基因扰动技术大多将基因功能简化为"开/关"的二元模式，难以捕捉剂量依赖的连续效 应。这种局限不仅削弱了对基因功能的全面理解，也可能在复杂生物学问题中引入不全面甚至错误的结论 。 2025 年 10 月 3 日，中国科学院深圳先进技术研究院合成生物学研究所 陈万泽 团队 与瑞士洛桑联邦理工 学院 Bart Deplancke 团队 合作，在 Nature 子刊 Nature Genetics 上 发表了题为： Dissecting the impact of transcription factor dose on cell reprogramming heterogeneity using scTF-seq 的研究论 文。 该研究开发了 scTF-seq （ 单细胞转录因子测序 ） 技术，实现了 剂量敏感的大规模基因扰动单细胞组学 ，并以转录因子介导的细胞重编程为模型，系统揭示了基因剂量在细胞命运调控中未被充分认识的多层 次、非线性复杂效应 。 编辑丨王多鱼 排版丨水成文 基因剂量的变化在胚胎发育、疾病发生以及细胞命运决定过程中发挥着关键作用，其系统化解析对于精准 理解生命过程至关重要。作为 ...

Core Viewpoint - The article discusses the groundbreaking Stereo-cell technology in single-cell sequencing, which overcomes the limitations of traditional methods by enabling multi-modal integration, real-time monitoring, and high-throughput capabilities, thus providing a comprehensive understanding of cellular characteristics and dynamics [4][20]. Group 1: Technology Breakthroughs - Stereo-cell technology achieves multi-modal integration, allowing for the simultaneous capture of cellular morphology, transcriptomics, and protein characteristics, akin to taking a "3D photo" of cells [12][13]. - The technology utilizes high-density DNA nanoball arrays, enabling unbiased capture of hundreds to millions of cells without the physical limitations of traditional methods, thus ensuring accurate identification of rare cell populations [11][9]. - Stereo-cell supports in-situ dynamic monitoring of cells, capturing gene transcription activity changes and spatial-temporal information, which expands the boundaries of single-cell research [15][19]. Group 2: Collaborative Initiatives - The establishment of the "10 Billion Cells Alliance" aims to create a comprehensive cell atlas and decode the underlying principles of life, driving innovation in life sciences from data accumulation to intelligent technology applications [4][20]. - The technology is expected to significantly impact clinical molecular medicine, providing new avenues for patient care and treatment through enhanced cellular analysis [20][21]. Group 3: Future Prospects - Stereo-cell is positioned as a next-generation life data engine, with plans to develop a "three major cell universe database" encompassing life maps, disease maps, and disturbance response maps, inviting global research teams to collaborate [20][22]. - The technology is anticipated to facilitate a transition from billions to trillions of cells in analysis, enabling a comprehensive understanding of cellular fates and functions throughout life [20].

Group 1: Historical Context of Pharmaceutical Development - The average human lifespan has significantly increased from the late 19th century, driven by advancements in the pharmaceutical industry [1] - The rise of modern drug development began in 19th century Europe, where the industrial revolution led to the use of coal tar for synthetic dyes, inadvertently paving the way for pharmaceutical innovations [2] - The establishment of germ theory by scientists like Pasteur and Koch revealed the causes of diseases, creating a market for drug development aimed at combating these diseases [3] Group 2: Evolution of Drug Discovery - The emergence of pharmacology as a distinct field in the late 19th century provided a systematic approach to drug discovery, connecting chemistry, biology, and clinical medicine [4] - The development of antibiotics, such as penicillin, marked a significant milestone in the pharmaceutical industry, allowing for the mass production of effective treatments against bacterial infections [4] - The continuous pursuit of interdisciplinary collaboration and deeper understanding of life mechanisms has been a driving force behind the progress in the pharmaceutical sector [5] Group 3: Challenges in the Pharmaceutical Industry - Despite advancements, the pharmaceutical industry faces high failure rates, with approximately 90% of new drug applications not receiving market approval [8] - The cost of developing new drugs has escalated, with the average cost rising by 145% from 2003 to 2013, reaching $2.6 billion [8] - The industry is under pressure due to the "high investment, low return" scenario, particularly in the treatment of chronic diseases [8] Group 4: The Role of AI in Pharmaceutical Innovation - The integration of AI and information technology is seen as a potential solution to the challenges faced by the pharmaceutical industry, enabling efficient analysis of genomic data and drug discovery processes [9] - AI models, such as AlphaFold, have revolutionized protein structure prediction, significantly enhancing the efficiency of drug design and reducing development costs [10] - AI-driven drug candidates are beginning to show promise in clinical trials, with companies like Insilico Medicine and Recursion advancing multiple drug candidates through various trial phases [12] Group 5: Future Prospects and Innovations - The concept of "virtual cells" and "digital twins" is emerging as a method to simulate human responses to drugs, potentially improving the accuracy of drug efficacy predictions [13] - The collaboration between various tech and research entities aims to leverage digital and AI technologies in drug design, potentially leading to new therapeutic categories [14] - While AI in drug development is still in its infancy, the potential for breakthroughs remains high, with ongoing research and investment driving the field forward [15]

Core Viewpoint - The article discusses the ambitious vision of creating Artificial Intelligence Virtual Cells (AIVC) to model and predict cellular behavior, leveraging advancements in AI and omics technologies [3][5][7]. Group 1: AI Virtual Cell Development - Multiple research teams are competing to develop AI models for cellular behavior prediction [4]. - The Chan-Zuckerberg Initiative (CZI) plans to invest hundreds of millions over the next decade to create virtual cells [10]. - The development of AI protein structure prediction tools like AlphaFold is contributing to virtual cell projects [10]. Group 2: Current Progress and Challenges - The efforts to create virtual cells are still in the early stages, generating significant interest in academic and industrial labs [8]. - Despite the excitement, some scientists express skepticism about the hype surrounding virtual cells, noting a lack of concrete results and clear success pathways [11]. - Current virtual cell models primarily focus on single-cell RNA sequencing data, which provides snapshots of gene activity and cellular states [16]. Group 3: Data Utilization and Future Directions - CZI plans to release sequencing data from 1 billion cells, while Arc Institute has released data from 100 million cancer cells treated with various drugs [16]. - Researchers are beginning to develop single-cell AI models, with Arc Institute launching its first virtual cell model called "State" [16]. - There is a need for integrating other data forms, such as optical and electron microscopy images, to enhance virtual cell models [17]. Group 4: Definition and Consensus - The concept of virtual cells lacks a clear definition, and there is no consensus among researchers on what constitutes a virtual cell [18]. - Stephen Quake emphasizes that the transition to using virtual cell models in biology will take time, as both the models and the scientists are not yet fully prepared [19].

Core Viewpoint - The article discusses the emergence of Virtual Cells (VC) as a frontier in the intersection of artificial intelligence and biology, aiming to revolutionize life sciences research by predicting cellular responses to disturbances [2][6]. Group 1: Virtual Cell Challenge - The Virtual Cell Challenge was launched by Arc Institute, with sponsorship from NVIDIA, 10x Genomics, and Ultima Genomics, offering cash prizes of $100,000, $50,000, and $25,000 for the top three models that accurately predict cellular responses to genetic disturbances [4]. - The challenge aims to provide a fair and open evaluation framework to identify the best virtual cell models through rigorous testing [2][4]. Group 2: Importance of Virtual Cells - Understanding and predicting cellular responses to disturbances, such as gene knockout or drug treatment, is a core challenge in biological and medical research [6]. - Advances in single-cell sequencing technology and breakthroughs in AI have reignited efforts to develop powerful virtual cell models that can predict responses across different cell types and states [6][20]. Group 3: Challenges in the Field - A significant bottleneck in the field is the lack of standardized evaluation criteria to assess whether a model truly understands cell biology rather than merely memorizing specific patterns in data [10]. - The Virtual Cell Challenge draws inspiration from the success of the CASP competition in protein structure prediction, which has catalyzed advancements in AI tools like AlphaFold [10]. Group 4: Challenge Design - The core task of the challenge is to assess the "cross-environment generalization" ability of models, requiring them to predict gene expression changes in a new cell type based on limited data from known cell types [13]. - A rigorous three-tier evaluation system is established to avoid model bias, focusing on differential expression scores, disturbance differentiation scores, and mean absolute error [14][15]. Group 5: Anticipated Impact - The challenge sets a benchmark for the industry by establishing a rigorous evaluation framework for predicting gene-level disturbance responses, guiding future developments in the field [19]. - It aims to promote data standards and reproducibility in single-cell functional genomics, accelerating the evolution of AI models through community competition and collaboration [19]. - The initiative is expected to gather global research efforts to tackle the challenges of virtual cell modeling, facilitating the transition from laboratory research to practical applications [19]. Group 6: Future Prospects - The first Virtual Cell Challenge focuses on gene disturbance predictions within a single cell type, with plans for future challenges to include combination disturbance predictions and integration of multi-omics data [20]. - The launch of the Virtual Cell Challenge signifies a new phase in AI-enabled life sciences, potentially transforming human understanding and intervention capabilities in biology [20].

Core Viewpoint - Xaira Therapeutics, an AI drug discovery company, has raised $1 billion in seed funding and aims to revolutionize drug discovery through advanced AI technologies [2] Group 1: Company Overview - Xaira Therapeutics was founded in April 2024 and has a distinguished leadership team, including Nobel laureates and former executives from major pharmaceutical companies [2] - The company has released the largest publicly available Perturb-seq dataset, named X-Atlas/Orion, which supports virtual cell research and enhances drug discovery predictions [3][4] Group 2: Dataset Details - The X-Atlas/Orion dataset includes 8 million cells and covers all protein-coding genes in humans, with over 16,000 unique molecular identifiers from single-cell deep sequencing [4][8] - Perturb-seq technology used in this dataset allows for the simultaneous reading of CRISPR sgRNA genetic perturbations and transcriptomes, revealing dose-dependent genetic effects [4][9] Group 3: Technological Innovations - The FiCS Perturb-seq platform developed by Xaira Therapeutics enables scalable production of perturbation sequencing data, overcoming previous limitations in data generation [8][11] - The platform demonstrates high sensitivity and minimal batch effects, accurately capturing transcriptional changes due to genetic perturbations [8] Group 4: Research Implications - The release of X-Atlas/Orion addresses the challenges of data generation scalability and standardization, facilitating the development of next-generation foundational models in life sciences [11] - The dataset will be shared under a non-commercial license to promote open collaboration in the biotechnology sector, with potential commercial partnerships available [12] Group 5: Future Directions - Xaira Therapeutics plans to expand data generation to include induced pluripotent stem cells and in vivo animal models, aiming for broader applications in AI-driven virtual cell development [20]