To Deconvolve, or Not to Deconvolve: Inferences of Neuronal Activities using Calcium Imaging Data

TL;DR

This paper compares direct calcium trace analysis and deconvolved spike train methods in neuroscience, finding that spike data generally outperform calcium traces in clustering, PCA, and decoding tasks, informing best practices.

Contribution

It provides a systematic comparison of calcium trace versus spike train analysis methods across multiple common neuroscience data analyses.

Findings

Spike data outperform calcium traces in clustering and PCA.

Calcium traces have higher predictability at individual time points.

Spike history or cumulative counts are better for population decoding.

Abstract

With the increasing popularity of calcium imaging data in neuroscience research, methods for analyzing calcium trace data are critical to address various questions. The observed calcium traces are either analyzed directly or deconvolved to spike trains to infer neuronal activities. When both approaches are applicable, it is unclear whether deconvolving calcium traces is a necessary step. In this article, we compare the performance of using calcium traces or their deconvolved spike trains for three common analyses: clustering, principal component analysis (PCA), and population decoding. Our simulations and applications to real data suggest that the estimated spike data outperform calcium trace data for both clustering and PCA. Although calcium trace data show higher predictability than spike data at each time point, spike history or cumulative spike counts is comparable to or better than calcium traces in population decoding.