人脑组织-机器生物计算机问世.pdf

112024 年 46 卷 1 期 特约专稿 Chinese Journal of Nature 2024 Vol.46 No.1 INVITED SPECIAL PAPERClimate change and carbon neutralityZHOU Tianjun,CHEN Xiaolong,ZHANG Wenxia,ZHANG LixiaState Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics,Institute of Atmospheric Physics,Chinese Academy of Sciences,Beijing 100029,ChinaAbstract Since the industrial revolution,carbon dioxide concentration in the atmosphere has risen sharply due to the widely-used fossil fuel,leading to global warming and frequent extreme climate events.To effectively confront and mitigate the climate change,in the Paris Agreement,the international community set warming targets to hold the increase in the global average temperature to well below 2 above pre-industrial levels and pursuing efforts to limit the temperature increase to 1.5 above pre-industrial levels.The concept of carbon neutrality starts spreading and known by larger communities.Here,we introduced some basic facts of global warming and the effect of anthropogenic carbon emission on the warming,summarized main results with respect to future climate changes,elucidated the relationship between 1.5/2 warming threshold and the carbon neutral,and finally overviewed the efforts of carbon reduction made by international community to effectively adapt to and mitigate the global warming.Key words climate change,carbon neutrality,extreme climate event,climate modeling,climate projection(编辑：沈美芳)人脑组织-机器生物计算机问世在2023年12月11日发表于 自然-电子学 的一项研究中，科学家制造了一种混合生物计算机，将实验室培养的人类脑组织与传统电子硬件结合在一起，可以完成语音识别等任务。这项技术有朝一日可能会被集成到人工智能系统中，或者成为神经科学研究中改进大脑模型的基础。研究人员将该生物计算机系统称为Brainoware。它结合了大脑类器官 一种模仿人体细胞的组织，用于研究器官模型。类器官由能够分化成不同类型细胞的干细胞制成。在这种情况下，它们变形为神经元，类似我们大脑中的神经元。论文通信作者、美国印第安纳大学伯明顿分校生物工程师郭峰表示，这项研究旨在建立“人工智能和类器官之间自然信息的桥梁”。人工智能和大脑都依赖于一个由互相连接的节点构成的网络传递信号，即所谓的神经网络。郭峰说：“我们想问一个问题是否可以利用大脑类器官内的生物神经网络进行计算？”为了制造Brainoware，研究人员将一个类器官放置在一块包含数千个电极的板上，将大脑与电路连接起来。然后，他们把想要输入的信息转换成一种电脉冲模式，并将其传递给类器官。脑组织的反应由传感器接收，并使用机器学习算法“解码”，该算法可以识别与之相关的信息。为了测试Brainoware的能力，研究人员使用这项技术进行了语音识别。方法是用8个人说话的240段录音进行训练，将音频转换为电信号并传递给类器官。Brainoware的迷你大脑对每种声音的反应都不同，进而产生了不同的神经活动模式。人工智能最终学会了通过解读这些反应来识别说话者。经过训练，该系统识别声音的准确率为78%。美国约翰斯霍普金斯大学发育神经学科学家Lena Smirnova说，尽管还需要更多研究，但这项研究证实了一些关键的理论观点，最终有望使生物计算机成为可能。之前的实验表明，只有二维培养的神经元细胞才能够完成类似的任务，而这是第一次在三维大脑类器官中实现这一功能。郭峰说，将类器官和计算机结合起来，可以让研究人员利用人脑的速度和能量效率发展人工智能。“这项技术可以用于研究大脑。”意大利比萨大学生物医学工程师Arti Ahluwalia说，因为大脑类器官可以复制真实活动的大脑结构和功能，而简单的细胞培养无法做到这一点。使用Brainoware模拟和研究神经系统疾病，如阿尔茨海默病，也是有潜 下转第34页34Chinese Journal of Nature 2024 Vol.46 No.1 REVIEW ARTICLEAdvances in structural biology of photosynthetic core complexes in purple bacteriaYUE Xingyu,WANG Zhengyu,YU Longjiang Key Laboratory of Photobiology,Institute of Botany,Chinese Academy of Sciences,Beijing 100093,China;University of Chinese Academy of Sciences,Beijing 100049,China;Faculty of Science,Ibaraki University,Mito 310-8512,JapanAbstract As an ancient model organism for studying molecular mechanisms of photosynthesis,purple bacteria convert solar energy into chemical energy through anoxygenic photosynthesis.The primary reaction of photosynthesis mainly occurs in the LH1-RC core complex,which is composed of light-harvesting antenna complex 1(LH1)and the reaction center(RC).Recently,structures of various bacterial LH1-RC complexes have been determined with developments of crystallography and cryo-EM techniques,significantly expanding our knowledge of this complex.In this paper,we systematically summarize the recent progress in structural biology research on the LH1-RC core complex from purple phototrophic bacteria,providing valuable insights for the deeper understanding of the molecular mechanisms of photosynthesis,improving photosynthetic efficiency,and advancing development and application of artificial photosynthetic systems.Key words photosynthesis,purple bacterium,light-harvesting complex,reaction center complex,LH1-RC core complex(编辑：段艳芳)59 COGDELL R J,GALL A,KHLER J.The architecture and function of the light-harvesting apparatus of purple bacteria:from single molecules to in vivo membranes J.Q Rev Biophys,2006,39:227-324.60 REIMERS J R,BICZYSKO M,BRUCE D,et al.Challenges facing an understanding of the nature of low-energy excited states in photosynthesis J.Biochim Biophys Acta,2016,1857:1627-1640.61 TUSCHAK C,BEATTY J T,OVERMANN J.Photosynthesis genes and LH1 proteins of Roseospirillum parvum 930I,a purple non-sulfur bacterium with unusual spectral properties J.Photosynth Res,2004,81:181-199.62 SAINI M K,CHIHCHE W,SOULIER N,et al.Caldichromatium japonicum gen.nov.,sp.nov.,a novel thermophilic phototrophic purple sulphur bacterium of the Chromatiaceae isolated from Nakabusa hot springs,Japan J.Int J Syst Evol Microbiol,2020,70:5701-5710.63 MADIGAN M T.Anoxygenic phototrophic bacteria from extreme environments J.Photosynth Res,2003,76:157-171.64 PERMENTIER H P,NEERKEN S,OVERMANN J,et al.A bacteriochlorophyll a antenna complex from purple bacteria absorbing at 963 nm J.Biochemistry,2001,40:5573-5578.65 TANI K,KANNO R,MAKINO Y,et al.Cryo-EM structure of a Ca2+-bound photosynthetic LH1-RC complex containing multiple-polypeptides J.Nat Commun,2020,11:4955.66 TANI K,KOBAYASHI K,HOSOGI N,et al.A Ca2+-binding motif underlies the unusual properties of certain photosynthetic bacterial core light-harvesting complexes J.J Biol Chem,2022,298:101967.67 NAMOON D,RUDLING N M,CANNIFFE D P.The role of the subunit in the photosystem of the lowest-energy phototrophs J.Biochemical Journal,2022,479:2449-2463.自然信息力的。通过观察类器官的反应，可以测试不同治疗方法的效果和副作用。“这就是希望所在。”Ahluwalia说，“有一天能用这些技术取代动物的大脑模型。”然而，使用活细胞进行计算的一大问题是如何使类器官长久存活。细胞必须在培养箱中生长和维护，而类器官越大，这一点就越难实现。更复杂的任务需要更大的“大脑”。郭峰表示，为了在Brainoware的能力基础上进一步发展，接下来的工作包括研究大脑类器官是否以及如何适应更复杂的任务，并将其设计得更稳定、更可靠。他说，如何把它们整合到目前用于人工智能计算的硅芯片中，这将是至关重要的。吴玉 编译 上接第11页