Modern neural recording techniques allow neuroscientists to obtain spiking activity from many hundreds of neurons simultaneously over long time periods, and new statistical methods are needed to understand structure of the large-scale data, in terms of both neuron numbers and recording duration. Here, we develop a bi-clustering method to cluster the neural spiking activity both spatially and temporally, according to their low-dimensional latent structures. The spatial (neuron) clusters are defined by the latent trajectories within each neural population, while the temporal (state) clusters are defined by local linear dynamics manner across the population. To flexibly extracting the bi-clustering structure, we build the model non-parametrically, and develop an efficient Markov chain Monte Carlo (MCMC) algorithm to sample the posterior distributions of model parameters. Validating our proposed MCMC algorithm through simulations, we find the method can recover unknown parameters and true bi-clustering structures successfully. We then apply the proposed bi-clustering method for counting series to multi-regional neural recordings under different experiment settings, where we find that simultaneously considering latent trajectories and spatial-temporal clustering structures can provide us with a more accurate and interpretable results. Overall, the proposed method provides scientific insights for large-scale (counting) time series with elongated recording periods, and it can have application beyond neuroscience.

翻译：现代神经记录技术使神经科学家能够同时获取数百个神经元在长时间内的尖峰活动，因此需要新的统计方法来理解大规模数据（包括神经元数量与记录时长）的结构。本文提出一种双聚类方法，根据低维潜在结构对神经尖峰活动进行空间和时间上的聚类。空间（神经元）聚类由每个神经群体内的潜在轨迹定义，而时间（状态）聚类则由群体间的局部线性动态方式定义。为灵活提取双聚类结构，我们采用非参数方式构建模型，并开发了一种高效的马尔可夫链蒙特卡洛（MCMC）算法对模型参数的后验分布进行采样。通过模拟验证该MCMC算法，发现该方法能成功恢复未知参数和真实双聚类结构。随后，我们将所提出的计数序列双聚类方法应用于不同实验设置下的多区域神经记录数据，发现同时考虑潜在轨迹与时空聚类结构可提供更准确且可解释的结果。总体而言，该方法为长记录周期的大规模（计数）时间序列提供了科学洞见，且可推广至神经科学以外的领域。