Towards a biophysical model of human cortical neurophysiological signatures that incorporates cellular and cell type biophysics, transcriptomics, and morphology

项目摘要 在过去的十年中，单细胞水平上的数据空前积累， 基因表达模式、形态学和电生理学特征的多模态集合。这么高- 对大脑中神经元类型的分辨率普查构成了迈向机械的重要一步。 理解电路和大脑计算。然而，将细胞身份和活动与电路和 神经生理信号仍然令人生畏。该提案的核心目标是制定一套公开的 可用的计算模型，工具和资源，允许测试关于角色的机械假设 不同神经元类型的神经回路功能，可用于分析神经数据，并可用于 确定大脑皮层功能的潜在生物标志物。建议的方法从建立 个体人类新皮层神经元的严格计算模型，将转录、形态和 physiology.接下来，我们将神经元连接到皮层回路中，模拟它们的活动，并模拟相关的 产生的神经生理信号在所有步骤中，模型都是由 现有的最先进的测量方法。作为原理证明，我们使用人类皮层电路模型， 研究由h通道携带的超极化激活的非特异性阳离子电流Ih的影响。 具体地说，我们最近表明，人类锥体FREM 3神经元内的Ih差异改变了它们的 突触整合特性。在FREM 3中沿沿着人体2/3层存在Ih深度梯度 事实上，神经元是颗粒上皮层的基本组织原则。即便如此，它的作用和它如何 影响电路处理和相关的神经生理信号仍然未知。人类大脑皮层 电路模型提供了一个理想的测试平台，其中关于FREM 3 Ih深度梯度作用的假设可以 得到考验首先，我们假设，虽然FREM 3模型的基本尖峰特性仍然广泛 不受深层与浅层FREM 3 Ih差异的影响，电路活动将受到实质性影响， 重合检测梯度将沿着皮质深度轴出现（由图1中的Ih梯度促进）。 FREM 3）。此外，我们假设，在FREM 3增加Ih和 在符合检测中，局部细胞外信号将在较慢的频带中表现出较低的功率，并且在较低的频带中表现出较高的功率。 更快的力量。因此，我们预计减少Ih（浅表神经元），导致相对增加， 在θ-慢伽马波段中的功率和在中高伽马波段中的减小。为了验证这些假设，我们将 在我们的电路模型中实例化FREM 3的Ih电导中的特定扰动，同时留下所有其他扰动。 模型参数不变。我们希望这些工具将与科学界分享， 将促进细胞和细胞类型特性，神经生理学， 测量和机械方法来阐明人类大脑回路中的功能。

Project Summary The past decade has seen an unprecedented accumulation of data at the single-cell level giving rise to unique, multimodal sets of gene expression patterns, morphologies, and electrophysiology characteristics. This high- resolution census of neuronal types in the brain constitutes an essential step towards a mechanistic understanding of circuits and brain computations. However, linking cellular identity and activity to circuits and neurophysiological signals remains daunting. The central goal of this proposal is to develop a set of publicly available computational models, tools and resources that allow testing mechanistic hypotheses about the role of different neuronal types in neural circuit function, that can be used to analyze neural data, and can serve to identify potential biomarkers of cortical function in the brain. The proposed approach starts with building rigorous computational models of individual human neocortical neurons linking transcription, morphology, and physiology. Next, we connect neurons into cortical circuits, emulate their activity and simulate the associated neurophysiological signals generated from those circuits. At all steps, the models are guided by, and compared to, existing state-of-the-art measurements. As a proof-of-principle, we use a human cortical circuit model to study the impact of the hyperpolarization-activated non-specific cation current, Ih, carried by h-channels. Specifically, we recently showed that Ih differences within human pyramidal FREM3 neurons alter their synaptic integration properties. The existence of an Ih depth gradient along human layer 2/3 in FREM3 neurons in fact is a fundamental organizing principle of the supragranular cortex. Even so, its role and how it affects circuit processing and associated neurophysiological signals remains unknown. The human cortical circuit model provides an ideal testbed where hypotheses about the role of the FREM3 Ih depth gradient can be tested. First, we hypothesize that while fundamental spike properties of FREM3 models remain broadly unaffected by deep vs. superficial FREM3 Ih-differences, the circuit activity will be substantially affected and a coincidence detection gradient will emerge along the cortical depth axis (facilitated by the Ih gradient in FREM3). Furthermore, we hypothesize that deep in the circuit where FREM3 has both increased Ih and coincidence detection, the local extracellular signals will exhibit lower power in the slower bands and increased power in faster ones. Accordingly, we expect decreased Ih (superficial neurons) to lead to a relative increase in power in theta-slow gamma and a decrease in the mid-high gamma band. To test these hypotheses, we will instantiate specific perturbations in the Ih conductance of FREM3 in our circuit model while leaving all other model parameters unperturbed. We expect that these tools, which will be shared with the scientific community, will facilitate the dialogue between the study of cellular and cell type properties, neurophysiological measurements, and mechanistic approaches to elucidate function in human brain circuits.