Drakopoulos Vasilios, Reichenbach Alex, Stark Romana, Foldi Claire J, Jean-Richard-Dit-Bressel Philip, Andrews Zane B
Monash Biomedicine Discovery Institute and Department of Physiology, Monash University, Clayton, Victoria 3800, Australia
Monash Biomedicine Discovery Institute and Department of Physiology, Monash University, Clayton, Victoria 3800, Australia.
eNeuro. 2025 Aug 14;12(8). doi: 10.1523/ENEURO.0221-25.2025. Print 2025 Aug.
Fiber photometry is a neuroscience technique that can continuously monitor in vivo fluorescence to assess population neural activity or neuropeptide/transmitter release in freely behaving animals. Despite the widespread adoption of this technique, methods to statistically analyze data in an unbiased, objective, and easily adopted manner are lacking. Various pipelines for data analysis exist, but they are often system specific, are only for preprocessing data, and/or lack usability. Current post hoc statistical approaches involve inadvertently biased user-defined time-binned averages or area under the curve analysis. To date, no post hoc user-friendly tool with few assumptions for a standardized unbiased analysis exists, yet such a tool would improve reproducibility and statistical reliability for all users. Hence, we have developed a user-friendly post hoc statistical analysis package in Python that is easily downloaded and applied to data from any fiber photometry system. This Fiber Photometry Post Hoc Analysis (FiPhoPHA) package incorporates a variety of tools, a downsampler, bootstrapped confidence intervals (CIs) for analyzing peri-event signals between groups and compared with baseline, and permutation tests for comparing peri-event signals across comparison periods. We also include the ability to quickly and efficiently sort the data into mean time bins, if desired. This provides an open-source, user-friendly Python package for unbiased and standardized post hoc statistical analysis to improve reproducibility using data from any fiber photometry system.
纤维光度法是一种神经科学技术，可在自由活动的动物体内连续监测荧光，以评估群体神经活动或神经肽/神经递质释放。尽管该技术已被广泛采用，但仍缺乏以无偏、客观且易于采用的方式对数据进行统计分析的方法。现有的各种数据分析流程往往是特定于系统的，仅用于数据预处理，和/或缺乏可用性。当前的事后统计方法涉及无意中存在偏差的用户定义的时间分箱平均值或曲线下面积分析。迄今为止，还没有一个假设少、适用于标准化无偏分析的用户友好型事后工具，但这样的工具将提高所有用户的可重复性和统计可靠性。因此，我们用Python开发了一个用户友好型的事后统计分析包，它易于下载并应用于来自任何纤维光度系统的数据。这个纤维光度法事后分析（FiPhoPHA）包包含了各种工具、一个下采样器、用于分析组间事件周围信号并与基线进行比较的自举置信区间（CI），以及用于比较不同比较期内事件周围信号的置换检验。如果需要，我们还具备快速有效地将数据分类到平均时间箱中的能力。这提供了一个开源、用户友好的Python包，用于无偏和标准化的事后统计分析，以使用来自任何纤维光度系统的数据提高可重复性。
