The GIFFittingToolbox is a Python-based repository designed for fitting and analyzing Generalized Integrate-and-Fire (GIF) neuron models. This repository is a fork of the original implementation, which was written in Python 2. It has been updated to Python 3 and extended for clustering applications for Joshi et. al. 2025. The toolbox provides tools for parameter estimation, model simulation, and evaluation of neuronal dynamics. The models are fitted on the following dataset:
Neuromodulatory influence on information transfer of single cortical neurons Joshi, N. Celikel, T. Zeldenrust, F. Kole, K. Safavi, S.P. Ak, A. Yan, X. Calcini, N. 2025. doi: https://doi.org/10.34973/nhes-qe97
The accompnying publication is the following:
Neuronal Identity is Not Static—An Input-Driven Perspective Nishant Joshi, Sven van Der Burg, Tansu Celikel, Fleur Zeldenrust bioRxiv 2024.10.16.618657; doi: https://doi.org/10.1101/2024.10.16.618657
The original repo is based on the following two publications:
C. Pozzorini, S. Mensi, O. Hagens, R. Naud, C. Koch and W. Gerstner, Automated high-throughput characterization of single neurons by means of simplified spiking neuron models,PLOS Computational Biology 2015
S. Mensi, O. Hagens, W. Gerstner and C. Pozzorini, Enhanced sensitivity to rapid input fluctuations by nonlinear threshold dynamics in neocortical pyramidal neurons, PLOS Computational Biology 2016
The repo also contains the 5 levels of Generalized Leaky Integrate and Fire models based on:
Teeter, C., Iyer, R., Menon, V. et al. Generalized leaky integrate-and-fire models classify multiple neuron types. Nat Commun 9, 709 (2018). https://doi.org/10.1038/s41467-017-02717-4
The fit is generated using BluePyOpt package optimized with Mutli-processing package for faster convergence.
Please look at Test_GLIF.py for examples.
- Parameter fitting for GIF models.
- Simulation of neuronal responses.
- Tools for analyzing, clustering, and visualizing results.
- Python 3.x
- Dependencies listed in
requirements.txt.
Refer to the documentation and examples in the src directory for guidance on using the toolbox.
This project is licensed under the MIT License.