TRACE: Contrastive learning for multi-trial time-series data in neuroscience

TL;DR

TRACE is a novel contrastive learning framework tailored for multi-trial neural time-series data, leveraging trial averaging to improve representation quality and capture biologically meaningful structures.

Contribution

It introduces a new contrastive learning method that exploits multi-trial data structure, outperforming existing approaches in neural data analysis.

Findings

TRACE outperforms other methods in simulated data

It captures biologically relevant continuous variation

It helps in data quality control

Abstract

Modern neural recording techniques such as two-photon imaging or Neuropixel probes allow to acquire vast time-series datasets with responses of hundreds or thousands of neurons. Contrastive learning is a powerful self-supervised framework for learning representations of complex datasets. Existing applications for neural time series rely on generic data augmentations and do not exploit the multi-trial data structure inherent in many neural datasets. Here we present TRACE, a new contrastive learning framework that averages across different subsets of trials to generate positive pairs. TRACE allows to directly learn a two-dimensional embedding, combining ideas from contrastive learning and neighbor embeddings. We show that TRACE outperforms other methods, resolving fine response differences in simulated data. Further, using in vivo recordings, we show that the representations learned by TRACE capture both biologically relevant continuous variation, cell-type-related cluster structure, and can assist data quality control.