eFEL: Electrophysiology Feature Extraction Library

Motivation Electrophysiological recordings are essential in experimental and computational neuroscience, providing insights into neuronal excitability and network behaviour. Extracting features such as action potential thresholds, widths, and firing patterns is conceptually straightforward, but in practice it is complicated by heterogeneous datasets and software environments, which hinder reproducibility and interoperability. A standardised, efficient, and portable framework is needed to ensure consistent analysis across platforms and alignment with community data standards.

Results We present the Electrophysiology Feature Extraction Library (eFEL), a cross-platform, open-source library that implements standardised definitions for over 90 electrophysiological features. eFEL combines a high-performance C ++ core with a Python interface, supporting customisable feature dependencies, caching, and parallelisation. It integrates with community standards such as NWB and works seamlessly with common electrophysiology formats and simulation environments. Since its initial release in 2015, eFEL has been used in published studies spanning single-cell analysis, model optimisation, multimodal fitting, and circuit simulations. eFEL provides a FAIR-compliant, versatile resource for reproducible electrophysiological data analysis.

Availability and implementation The eFEL library is publicly available at https://github.com/openbraininstitute/eFEL and the associated study data and scripts have been deposited in Zenodo at https://zenodo.org/records/17241835. Supplementary information Supplementary data are available at Bioinformatics online.