We publish two electrophysiological datasets recorded in motor cortex of two macaque monkeys during an instructed delayed reach-to-grasp task, using chronically implanted 10-by-10 Utah electrode arrays. We provide a) raw neural signals (sampled at 30 kHz), b) time stamps and spike waveforms of offline sorted single and multi units (93/49 and 156/19 SUA/MUA for the two monkeys, respectively), c) trial events and the monkey’s behavior, and d) extensive metadata hierarchically structured via the odML metadata framework (including quality assessment post-processing steps, such as trial rejections). The dataset of one monkey contains a simultaneously saved record of the local field potential (LFP) sampled at 1 kHz. To load the datasets in Python, we provide code based on the Neo data framework that produces a data structure which is annotated with relevant metadata. We complement this loading routine with an example code demonstrating how to access the data objects (e.g., raw signals) contained in such structures. For Matlab users, we provide the annotated data structures as mat files.
展开▼
机译：我们使用慢性植入的10×10犹他州电极阵列，在指令的延迟到达抓握任务期间，发布了记录在两只猕猴的运动皮质中的两个电生理数据集。我们提供a）原始神经信号（以30 kHz采样），b）离线分类的单个和多个单位的时间戳和尖峰波形（分别为两只猴子的93/49和156/19 SUA / MUA），c）试用事件和猴子的行为，以及d）通过odML元数据框架按层次结构构建的广泛元数据（包括质量评估后处理步骤，例如试验拒绝）。一只猴子的数据集包含同时保存的以1 fieldkHz采样的局部场电势（LFP）的记录。为了用Python加载数据集，我们提供了基于Neo数据框架的代码，该框架生成了一个数据结构，并用相关的元数据进行了注释。我们用示例代码来补充此加载例程，该示例代码演示如何访问此类结构中包含的数据对象（例如原始信号）。对于Matlab用户，我们提供带注释的数据结构作为mat文件。
展开▼
