A. Mathis Group

There is a newer version of the record available.

Published February 18, 2024 | Version v1
Software Open

Task-driven neural network models predict neural dynamics of proprioception: Code

Description

#############

Task-driven neural network models predict neural dynamics of proprioception, BioRxiv 2023

#############

Authors: Marin Vargas, Alessandro (orcid=0000-0001-7073-4120) and Bisi, Axel (orcid=0009-0006-8602-7555) and Chiappa, Alberto Silvio (orcid=0009-0001-2764-6552) and Versteeg, Christopher (orcid=0000-0002-4269-5109) and Miller, Lee E. (orcid=0000-0001-8675-7140) and Mathis, Alexander (orcid=0000-0002-3777-2202)

Affiliation: EPFL

Date: January, 2024

Link to the BioRxiv article: 

https://www.biorxiv.org/content/10.1101/2023.06.15.545147v1.abstract

--------------------------------

Here we provide the code of our article "Task-driven neural network models predict neural dynamics of proprioception". It contains the pipeline to generate the synthetic training dataset of simulated muscle spindles, the scripts for training neural network models and for predicting the neural activity.

The overall structure of the data is:

├── PCR-data-generation           - Contains the scripts to generate the synthetic proprioceptive dataset
├── rl-data-generation               - Contains the scripts to convert RL pybullet data to muscle spindle
├── code                                     - Contains the utils for training neural network models and paths
├── nn-training                           - Contains the scripts to train neural network models of proprioception
├── exp_data_processing            - Contains the scripts to process the experimental data
├── neural_prediction                 - Contains the scripts to use pre-trained neural networks to predict experimental neural data
└── paper_figures                       - Contains the scripts to generate main figures of the paper

The data to reproduce the results is available at:
- Synthetic muscle spindles dataset:                                                                   https://zenodo.org/records/10530013
- Weights of neural network models:                                                                  https://zenodo.org/records/10533584
- Experimental data, activations and predictions of neural network models:     https://zenodo.org/records/10542311
 
--------------------------------

Here we archive v1.0.  The code is released under the MIT license, for updates see: https://github.com/amathislab/Task-driven-Proprioception

If you find our code, weights, predictions or ideas useful, please cite:

@article{vargas2023task,
  title={Task-driven neural network models predict neural dynamics of proprioception},
  author={{Marin Vargas}, Alessandro and Bisi, Axel and Chiappa, Alberto Silvio and Versteeg, Christopher and Miller, Lee E and Mathis, Alexander},
  journal={bioRxiv},
  pages={2023--06},
  year={2023},
  publisher={Cold Spring Harbor Laboratory}
}

Files

readme.md

Files (2.8 kB)

Name Size Download all
md5:d2fda864ac6dcc0f3669ae01fdb54b2c
2.8 kB Preview Download

Additional details

Identifiers

DOI
10.1101/2023.06.15.545147

Funding

Swiss National Science Foundation
A theory-driven approach to understanding the neural circuits of proprioception 212516

Software

Programming language
Python