科学范式决定了科学组织模式。大数据驱动下，生命科学领域发生革命和范式转换。以往研究多以物理学为对象，认为科层制是科学的组织模式，而少有对生命科学组织的具体研究。研究的目的是为了探究与物理学不同的生命科学的组织模式，寻找大数据驱动下生命科学组织变革的动因，生命科学领域的组织机构和结构的变化，及生命科学知识生产模式的改变。研究运用了历史的方法、分析与综合等方法对大数据、生命科学革命、新范式、组织及组织模式等基本概念做了简要梳理。研究结合了范式理论、组织变革理论、知识生产模式理论和科学知识社会学理论，分析了生命科学组织变革的动因和主要变量要素，探究组织机构变化和结构改观，揭示了变革中知识生产模式的变化及特点，并给出了组织未来发展的建议。

研究得出：在大数据驱动下，第一，生命科学组织变革的外部驱动因素中政策导向、经济因素、社会文化认同和高新技术水平发挥了作用，加之生命科学的几次大事件如人类基因组计划、抗癌计划等改变了组织的任务目标。生命科学的内部驱动因素为分布式大科学装置、高的资产专用性和高的可配置的人力资本、信息化和高新技术的参与，以及贯穿科研流程的数据存储与管理、挖掘和分析、应用和转化的数据分析成为内驱性需求。

第二，从生命科学组织内容要素来看，生命科学组织的任务目标、科学方法、科学家角色和组织文化分别发生变革。生命科学的研究任务由假设、实验和问题驱动为主，发生转向数据驱动的“迭代”发现。在方法论上，数学一直作为工具参与科研，数法尤其是建模的据方补充使科研产生数据、做出科学发现，并向应用成果转化。生命科学家拥有多学科身份和与人工智能双重发现的主体。未来生命科学人才培养需要会使用高新技术、具备政治素养和经济头脑、会社会交往和活动的素质，在课程体系上《计算生物学》和《生物信息学》变得重要。从近代以来生命科学学科演化来看，变现出会聚的组织文化和策略。

第三，在生命科学组织的结构上，从英国组织社会学家惠特利的研究中可以发现，以生物医学、人工智能和工程学等为代表的专业动态组织与以物理学等、化学等分别代表的两种科层制之间有所不同，且专业动态组织有独特特点。在众包模式下，从科学公开赛中可见，公民科学家、企业机构、社区组织等参与者“涌现”。其中，以华大基因为代表的新型企业机构影响力突出。华大拥有自主研发的高通量测序仪、依托深圳国家数据库，通过合作高速、高存储的计算设备，并且以创新教育模式，校企合作办学，取得了领先科研和育人水平。无创产前基因检测为代表的医学成果也在国内国际广泛开展。在世界范围内，生命科学组织平台繁荣发展，企业集群会聚。生命科学组织结构呈现出继承了专业动态组织发展模式，又有平台化、生态化、集群会聚的特点，表现出共享化的研究方式。

第四，在知识生产模式Ⅰ和Ⅱ的基础上，基于生物大数据独性及应对，生命科学知识生产体现出“会聚-解聚”的知识生产阶段，其医学应用转化是为“愿景”驱动的基础研究。在科学知识共生产模式下，大数据驱动下的生命科学知识还具有不完备性、由因果性到相关性、共享性、自明性、应用性的特点。

第五，研究建议生命科学组织的未来发展在组织技术上，要打造跨部门的交流环境、促进基础设施建设、促进数据标准化建设；在组织设计上，可以以会聚观为指导，促进数据资源开放共享；在组织管理上，要借鉴经验，做分层级管理，发挥产业集群效应，成立科学中心，以及实现生命数据全周期管理，注意伦理、法律和社会问题。研究猜想生命科学和技术的革命将带来未来世界图景的变化，表现为数字化社会文明形态，带来人类自然观、生存方式和自我认知的变革。

The scientific paradigm determines the scientific organization model. Driven by big data, there has been a revolution and paradigm shift in the field of life sciences. In the past, most of the studies focused on physics, and considered the bureaucratic system to be a scientific organization model, but there were few specific studies on the organization of life sciences. The purpose of this study is to explore the organizational model of life sciences that are different from physics, and to find out the driving forces of life science organizational change driven by big data, the changes in the organizational structure and structure of life sciences, and the changes in life science knowledge production models. The research uses historical methods, analysis and synthesis methods to briefly sort out the basic concepts of big data, life science revolution, new paradigms, organizations and organizational models. This study combines paradigm theory, organizational change theory, knowledge production mode theory and scientific knowledge sociology theory, analyzes the motivation and main variable elements of life science organizational change, explores the change of organizational structure and structural change, reveals the changes and characteristics of knowledge production mode in the change, and gives suggestions for the future development of the organization.

The results show that, driven by big data, first, policy orientation, economic factors, social and cultural identity and high-tech level play a role in the external driving factors of life science organizational change, and several major events in life sciences, such as the Human Genome Project and the Anti-Cancer Project, have changed the mission goals of the organization. The internal driving factors of life sciences are distributed large scientific devices, high asset specificity and high configurable human capital, informatization and high-tech participation, as well as data analysis that runs through the scientific research process of data storage and management, mining and analysis, application and transformation.

Second, from the perspective of the content elements of life science organizations, the mission objectives, scientific methods, roles of scientists, and organizational culture of life science organizations have changed respectively. The research tasks of life sciences are mainly hypothesis, experimental, and problem-driven, and there is a shift to data-driven "iterative" discovery. In terms of methodology, mathematics has always been used as a tool to participate in scientific research, and mathematical methods, especially the basis for modeling, make scientific research generate data, make scientific discoveries, and transform applied results. Life scientists have a multidisciplinary identity and a subject of dual discovery with artificial intelligence. In the future, the cultivation of life science talents needs to be able to use high technology, have political literacy and economic acumen, and be able to communicate and carry out social activities, and "Computational Biology" and "Bioinformatics" have become important in the curriculum system. From the perspective of the evolution of life science disciplines in modern times, the organizational culture and strategy of convergence have emerged.

Third, in terms of the structure of life science organization, it can be found from the research of British organizational sociologist Richard that there are differences between the professional dynamic organization represented by biomedicine, artificial intelligence and engineering and the two discipline hierarchical systems represented by physics and chemistry, and the professional dynamic organization has unique characteristics. In the crowdsourcing model, it can be seen from the Science Open that citizen scientists, business institutions, community organizations and other participants "emerge". Among them, the new enterprise represented by BGI has outstanding influence. BGI has a self-developed high-throughput sequencer, relying on the Shenzhen national database, through cooperation with high-speed, high-storage computing equipment, and innovative education mode, school-enterprise cooperation in running schools, and has achieved a leading level of scientific research and education. The medical achievements represented by non-invasive prenatal genetic testing have also been widely carried out at home and abroad. Around the world, life sciences organizations are thriving and clusters of companies are converging. The organizational structure of life sciences presents the characteristics of inheriting the professional dynamic organizational development model, and has the characteristics of platformization, ecology and cluster convergence, showing a shared research method.

Fourth, on the basis of knowledge production models I and II, based on the uniqueness and response of biological big data, life science knowledge production reflects the knowledge production stage of "convergence and depolymerization", and its medical application transformation is a basic research driven by "vision". Under the mode of co-production of scientific knowledge, life science knowledge driven by big data also has the characteristics of incompleteness, from causality to relevance, sharing, self-evidentness and application.

Fifth, the study suggests that the future development of life science organizations should create a cross-departmental communication environment, promote infrastructure construction, and promote data standardization in terms of organizational technology; in terms of organizational design, we can promote the open sharing of data resources under the guidance of convergence view; in terms of organizational management, we should learn from experience, do hierarchical management, give full play to the effect of industrial clusters, establish science centers, and realize the full-cycle management of life data, and pay attention to ethical, legal and social issues. Research conjectures that the revolution in life sciences and technology will bring about changes in the future world picture, which will be manifested in the form of digital social civilization and bring about changes in human perceptions of nature, living methods and self-perception.