Regenerative Bioscience Center, University of Georgia, Athens, GA, USA.
Division of Neuroscience, Biomedical Health and Sciences Institute, University of Georgia, Athens, GA, USA.
Sci Rep. 2021 Apr 27;11(1):9110. doi: 10.1038/s41598-021-88675-w.
Microelectrode arrays (MEAs) are valuable tools for electrophysiological analysis, providing assessment of neural network health and development. Analysis can be complex, however, requiring intensive processing of large data sets consisting of many activity parameters, leading to information loss as studies subjectively report relatively few metrics in the interest of simplicity. In screening assays, many groups report simple overall activity (i.e. firing rate) but omit network connectivity changes (e.g. burst characteristics and synchrony) that may not be evident from basic parameters. Our goal was to develop an objective process to capture most of the valuable information gained from MEAs in neural development and toxicity studies. We implemented principal component analysis (PCA) to reduce the high dimensionality of MEA data. Upon analysis, we found the first principal component was strongly correlated to time, representing neural culture development; therefore, factor loadings were used to create a single index score-named neural activity score (NAS)-reflecting neural maturation. For validation, we applied NAS to studies analyzing various treatments. In all cases, NAS accurately recapitulated expected results, suggesting viability of NAS to measure network health and development. This approach may be adopted by other researchers using MEAs to analyze complicated treatment effects and multicellular interactions.
微电极阵列（MEA）是电生理分析的有价值的工具，可评估神经网络的健康和发育情况。然而，分析可能很复杂，需要对由许多活动参数组成的大型数据集进行密集处理，这会导致信息丢失，因为研究人员为了简单起见，主观地报告相对较少的指标。在筛选试验中，许多小组报告简单的整体活动（即发射率），但忽略了网络连接性变化（例如爆发特征和同步性），这些变化可能无法从基本参数中明显看出。我们的目标是开发一种客观的过程，以捕获从神经发育和毒性研究中的 MEA 获得的大部分有价值的信息。我们实施了主成分分析（PCA）来降低 MEA 数据的高维性。经过分析，我们发现第一主成分与时间强烈相关，代表神经培养的发育；因此，因子负荷用于创建一个单一的指数得分-命名为神经活动得分（NAS）-反映神经成熟度。为了验证，我们将 NAS 应用于分析各种处理的研究。在所有情况下，NAS 准确地再现了预期的结果，这表明 NAS 可以用于测量网络的健康和发育。使用 MEA 进行分析的其他研究人员可以采用这种方法来分析复杂的处理效果和多细胞相互作用。
