Abstract
BACKGROUND: Chronic extracellular recordings are a powerful tool for systems neuroscience, but spike sorting remains a challenge. A common approach is to fit a generative model, such as a mixture of Gaussians, to the observed spike data. Even if non-parametric methods are used for spike sorting, such generative models provide a quantitative measure of unit isolation quality, which is crucial for subsequent interpretation of the sorted spike trains. NEW METHOD: We present a spike sorting strategy that models the data as a mixture of drifting t-distributions. This model captures two important features of chronic extracellular recordings-cluster drift over time and heavy tails in the distribution of spikes-and offers improved robustness to outliers. RESULTS: We evaluate this model on several thousand hours of chronic tetrode recordings and show that it fits the empirical data substantially better than a mixture of Gaussians. We also provide a software implementation that can re-fit long datasets in a few seconds, enabling interactive clustering of chronic recordings. COMPARISON WITH EXISTING METHODS: We identify three common failure modes of spike sorting methods that assume stationarity and evaluate their impact given the empirically-observed cluster drift in chronic recordings. Using hybrid ground truth datasets, we also demonstrate that our model-based estimate of misclassification error is more accurate than previous unit isolation metrics. CONCLUSIONS: The mixture of drifting t-distributions model enables efficient spike sorting of long datasets and provides an accurate measure of unit isolation quality over a wide range of conditions.