Scalable tools for consistent identification of neuronal cell types in mouse and human

用于一致识别小鼠和人类神经元细胞类型的可扩展工具

• 批准号: 10365216
• 负责人: Staci A Sorensen
• 金额: $ 108.14万
• 依托单位: ALLEN INSTITUTE
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-09-16 至 2024-09-15
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10365216
• 关键词: Acute; Address; Agreement; Algorithms; Anatomy; Axon; Biological; Brain; Brain imaging; Brightfield Microscopy; Cells; Characteristics; Collection; Communities; Computer software; Computing Methodologies; Coupled; Data; Data Set; Descriptor; Effectiveness; Electrophysiology (science); Functional disorder; Funding; Genetic Transcription; Gold; Human; Image; Individual; Institutes; Joints; Label; Learning; Literature; Location; Manuals; Measurement; Methods; Modality; Modernization; Molecular; Molecular Profiling; Morphology; Mus; Nervous system structure; Neurons; Neurosciences; Performance; Phenotype; Physiological; Population; Preparation; Property; Quality Control; Reproducibility; Research; Research Personnel; Science; Shapes; Slice; Software Tools; Speed; Supervision; Testing; Time; Training; Translating; Uncertainty; Vision; Visualization; Work; artificial neural network; autoencoder; automated analysis; automated segmentation; base; brain dysfunction; cell type; cloud based; cloud platform; computational pipelines; computerized tools; cost; data repository; experimental study; flexibility; image archival system; improved; innovation; large datasets; light microscopy; machine learning algorithm; machine learning method; microscopic imaging; multimodality; multiple datasets; neural network; patch sequencing; programs; reconstruction; success; tool; transcriptomics

Project Summary The proposed work will address a critical gap in our understanding of neuronal phenotypes and cell types by developing machine learning algorithms and cloud-based software for the integration of multiple modality characterizations large and growing datasets of cortical neurons in mouse and human. Through optimal and innovative use of potentially incomplete data and emphasis on automated morphological characterization, the proposed tools will enable richer and consistent characterizations of neurons from transcriptomic, anatomical, or electrophysiological profiling. While large-scale, BICCN-funded cell type research programs rely on the notion of unique neuronal identity which determines the cell’s phenotype across different observation modalities, overarching agreements across physiological, anatomical and molecular characterizations remain elusive. Although these large-scale programs succeeded in generating extensive multiple modality datasets, the lack of principled, accurate and widely available computational alignment and inference tools presents a roadblock to the success of the overall program. A second issue is that anatomical characterization, despite being the classical approach to understanding cell types, lags significantly behind molecular and physiological methods in terms of throughput. The research proposed here aims to address the alignment problem by building on the coupled autoencoder approach, which presents an efficient optimization framework centered on the ubiquity of neuronal identity. Importantly, the proposed software can utilize incompletely characterized data points, which is common in practice, to produce unified visualization and analysis of abstract neuronal identity. This tool will be both flexible (e.g., the feature set can be changed) and extensible (e.g., more observation modalities can be added for joint alignment). The aligned representations enable consistent clustering of the neuronal population across the different observation modalities, which is a pressing problem in modern neuroscience. We propose to address the anatomical throughput issue with an end-to-end computational pipeline, from the raw image of local neuronal arbors to the anatomical descriptor that can be readily aligned and interpreted by the coupled autoencoder software. By utilizing our extensive gold-standard manual reconstructions, we will train supervised deep artificial neural networks to segment neuronal arbors in sparse labeling scenarios. The rich set of training examples, together with algorithmic innovations, will endow superior generalizability of this automated segmentation tool, accelerating science for light microscopy-based studies.

项目概要 拟议的工作将通过以下方式解决我们对神经元表型和细胞类型的理解中的一个关键差距 开发机器学习算法和基于云的软件以集成多种模式 表征小鼠和人类皮层神经元的大型且不断增长的数据集。通过最优和 创新地使用可能不完整的数据并强调自动化形态表征， 所提出的工具将使神经元从转录组、解剖学或 电生理分析。 虽然 BICCN 资助的大规模细胞类型研究项目依赖于独特神经元身份的概念 它决定了细胞在不同观察模式下的表型，跨不同观察模式的总体协议 生理、解剖和分子特征仍然难以捉摸。虽然这些大型节目 成功地生成了广泛的多模态数据集，但缺乏原则性、准确性和广泛性 可用的计算对齐和推理工具为整体成功提供了障碍 程序。第二个问题是解剖学特征，尽管是经典的方法 了解细胞类型在通量方面明显落后于分子和生理方法。 这里提出的研究旨在通过耦合自动编码器来解决对齐问题 方法，它提出了一个以普遍存在的神经元身份为中心的有效优化框架。 重要的是，所提出的软件可以利用不完全表征的数据点，这在 实践，产生抽象神经元身份的统一可视化和分析。这个工具既灵活又灵活 （例如，可以更改特征集）和可扩展（例如，可以为联合添加更多观察模式） 结盟）。对齐的表示可以实现神经元群体的一致聚类 不同的观察方式，这是现代神经科学的一个紧迫问题。 我们建议通过端到端计算管道来解决解剖吞吐量问题，从原始 局部神经元乔木的图像与解剖描述符可以很容易地对齐和解释 耦合自动编码器软件。通过利用我们广泛的黄金标准手动重建，我们将训练 监督深度人工神经网络在稀疏标记场景中分割神经元乔木。丰富的套装 训练示例的数量以及算法创新将赋予这种自动化的卓越的通用性 分割工具，加速基于光学显微镜的研究。