Reverse-engineering Drosophila's retinal networks

视觉是许多生物最重要的感觉。即使在最昏暗的栖息地，动物也有功能正常的眼睛，这使它们能够从环境中提取有用的光学信息，并对不断变化的事件和条件做出快速和适当的反应。以速度、灵敏度、动态范围和稳健性衡量的生物体视觉系统的解剖结构、分子信号级联以及最终的性能，已经通过光刺激的内容和其他环境因素进行了调整，以适应其生活方式。无脊椎动物物种的视觉是广泛研究的主题，导致发现了适用于所有感官的重要基本原则。经过数十年的遗传学、解剖学、生理学和行为学研究，果蝇视觉系统是研究早期神经加工机制的理想模型。果蝇第一视觉突触层的超微结构已从电子显微镜切片上得到了完整的描述。大量的分子遗传学方法结合生化分析和细胞内记录技术已经汇聚到果蝇的视觉系统上，提供了前所未有的实验可操作性。这些方法使确定其信令级联的要素成为可能，并提供了对相关监管机制的独特见解。苍蝇光感受器的信号处理能力惊人，在许多方面都超过了人类工程图像传感器。它们非常敏感，能够对单光子事件做出反应。嵌入在噪声中的弱输入信号可以被选择性地放大和过滤，以提供对生理相关刺激的有效和可靠的检测。光感受器的一个迷人的功能属性尚未在工程传感设备中复制，即它们能够根据过去和正在进行的光事件，使用多层正反馈和负反馈控制来适应光，即调整放大增益。这使得它们可以在广泛的环境范围内运行。它们可以在黑暗适应的条件下可靠地对单光子的吸收做出反应，但也可以调整增益以在明亮的白天条件下工作。这个项目旨在利用从控制和系统工程中借用的数学工具和技术，开发一个早期视觉系统的详细数学模型，使我们能够了解有助于将光转换为电信号的不同分子组件的作用，以及飞行光感受器必须遵守的适应规则，以便在黑暗和充分的阳光下可靠地运行。

Vision is the most important sense of many living organisms. Even in the dimmest habitats animals have functional eyes, which allow them to extract useful optical information from the environment and to respond rapidly and appropriately to changing events and conditions. The anatomical structure, molecular signalling cascades and ultimately the performance of an organism's visual system, measured in terms of speed, sensitivity, dynamic range and robustness, has been tuned to suit its lifestyle by the content of the photic stimuli as well as other environmental factors. Vision of invertebrate species was the subject of extensive research which led to the discovery of important fundamental principles that apply to all senses. Drosophila visual system is an ideal model to investigate the mechanisms underlying early neural processing with a wealth of detailed knowledge from tens of years of genetic, anatomical, physiological and behavioural studies. The ultrastructure of the first visual synaptic layer in Drosophila has been fully described from electron-micrograph sections. A vast array of molecular genetic methods combined with biochemical analysis and intracellular recording techniques have converged on Drosophila's visual system providing unprecedented experimental tractability. These methodologies have made possible the identification of elements of its signalling cascades and offered unique insight into the associated regulatory mechanisms. The signal-processing capability of fly photoreceptors is prodigious, outperforming human engineered image sensors in many respects. They are exquisitely sensitive, being able to respond to single photon events. Weak input signals embedded in noise can be selectively amplified and filtered to provide efficient and reliable sensing of physiologically relevant stimuli. A fascinating functional attribute of photoreceptors, that is yet to be replicated in an engineering sensing device, is their ability to light adapt, i.e. adjust the amplification gain, according to both past and on-going light events, using many layers of positive- and negative-feedback control. This allows them to operate over a wide environmental range. They can reliably respond to the absorption of single photons under dark-adapted conditions, but can also adjust the gain to operate in bright daylight conditions. This project aims to develop, using mathematical tools and techniques borrowed from control and systems engineering, a detailed mathematical model the early vision system that will allow us to understand the role of different molecular components that are instrumental in converting light into electrical signals and the adaptation rules that fly photoreceptors must obey in order to operate reliably in dark as well as in full shinshine.