🧠 Functional Connectomes

This repository contains a Jupyter Notebook titled FunctionalConnectomes.ipynb, focusing on the analysis and modeling of functional connectomes within neurobiological networksFunctional connectomes represent the dynamic patterns of neural interactions and are crucial for understanding brain function and behavior

📖 Overview

The notebook provides an unsupervised approach to learn dynamic affinities between neurons in live, behaving animal. It employs pairwise non-linear affinities between neuronal traces from brain-wide calcium activity, organized by non-negative tensor factorization (NTF. Each factor specifies groups of neurons most likely interacting during inferred intervals, facilitating the revelation of dynamic functional connectome. This method allows for the identification of neural motifs active during different experimental stages, such as stimulus application or spontaneous behavior.

✨ Features

• *Dynamic Affinity Learning: Utilizes unsupervised learning techniques to capture time-varying neuronal interaction.
• *Non-negative Tensor Factorization (NTF): Applies NTF to organize neuronal activity data, highlighting functional motif.
• *Community Detection: Implements weighted community detection to infer dynamic functional connectome.

🚀 Getting Started

To explore the analyses and models presented:

1. Clone the Repository:

   git clone https://github.com/joaommata/Functional-Connectomes.git
   ``
   
   
   

2. Install Dependencies: Ensure you have Python and Jupyter Notebook installed. Install necessary packages using:

   pip install -r requirements.txt
   ``
   
   
   

3. Run the Notebook: Navigate to the repository directory and launch Jupyter Notebook:

   jupyter notebook FunctionalConnectomes.ipynb
   ``
   
    
   

🛠️ Usae

The notebook is structured to guide you through the process of loading neuronal activity data, applying NTF, and interpreting the resulting functional connectoe. Detailed explanations and code cells are provided for each sep.

🤝 Contributng

Contributions are weloe. Please fork the repository and submit a pull request with your enhancemnts.

📄 Licnse

This project is licensed under the MIT Liense.

🙏 Acknowledgents

This work is inspired by methodologies for learning dynamic representations of functional connectomes in neurobiological newors.

For more information, refer to the associated researchpaper: