Obtaining artifact-corrected signals in fiber photometry via isosbestic signals, robust regression, and dF/F calculations

Significance: Fiber photometry is a powerful tool for neuroscience. However, measured biosensor signals are contaminated by various artifacts (photobleaching and movement-related noise) that undermine analysis and interpretation. Currently, no universal pipeline exists to deal with these artifacts. Aim: We aim to evaluate approaches for obtaining artifact-corrected neural dynamic signals from fiber photometry data and provide recommendations for photometry analysis pipelines. Approach: Using simulated and real photometry data, we tested the effects of three key analytical decisions: choice of regression for fitting isosbestic control signals onto experimental signals [ordinary least squares (OLS) versus iteratively reweighted least squares (IRLS)], low-pass filtering, and dF/F versus dF calculations. Results: IRLS surpassed OLS regression for fitting isosbestic control signals to experimental signals. We also demonstrate the efficacy of low-pass filtering signals and baseline normalization via dF/F calculations. Conclusions: We conclude that artifact-correcting experimental signals via low-pass filter, IRLS regression, and dF/F calculations is a superior approach to commonly used alternatives. We suggest these as a new standard for preprocessing signals across photometry analysis pipelines.