HME-Toolbox

It's a C++ implementation of Hierarchical Mixture of Experts we've used in Bioinformatics & Cardiology

OVERVIEW

---

CNetSOM is a C++ implementation of a Self-Organizing Map (SOM) designed for clustering and classification tasks in bioinformatics/cardiology. This software is capable of processing training data to generate gene clusters, providing valuable insights into patterns within biological data.

Features

---

Self-Organizing Map (SOM): Implements a neural network model that organizes data into a grid based on similarity. Multiple Distance Metrics: Supports various distance calculations including Euclidean, Manhattan, and Pearson distances. Dynamic Node Expansion: Capable of dynamically expanding the network based on the training data. Classification Performance Evaluation: Evaluates the performance of classification against known labels. Output Generation: Outputs the generated gene clusters to files with .gdat extension.

Requirements

---

C++ Compiler (e.g., g++11 snd above) Standard C++ Libraries

Organization

The code is structured as:

project/

Blocks	Components	Description
src	main.cpp	# Main application file
	NNPattern.cpp	Implementation of the NNPattern class
	TNetSOM.cpp	Implementation of the TNetSOM class
include	NNPattern.h	Header for NNPattern class
	TNetSOM.h	Header for TNetSOM class
docs	README	Papers to use in your citations 4 code`

Compilation Instructions

To compile the project, follow these steps:

1. Open a terminal and navigate to the project directory.

2. Compile the source files using the following command:

   g++ -I include/ src/main.cpp src/NNPattern.cpp src/TNetSOM.cpp -o NetSOM

## Running the Program
----------------------
After compiling, you can run the program with the following command:

bash:
./NetSOM <data_file> <genes_exist>
<data_file>: Path to your input training data file.
<genes_exist>: Pass 1 if gene names are specified in your data file; otherwise, pass 0.

Example:
--------
./NetSOM training_data.txt 1

## OUTPUT:

After running the program, it will output generated gene clusters to files with a .gdat extension. For example, NodePatterns2043.gdat will contain the genes for node 20,43.

Additional Information:
* Ensure you have g++ installed on your system.
* Modify any parameters in the source files as needed for your specific use case.
* Check the header files for details on available methods and classes.

For any issues or contributions, please open an issue or submit a pull request.

Contributing
------------
Contributions are welcome! Please fork this repository and submit your pull requests for any improvements or bug fixes.

LICENSE
--------
This project is licensed under the MIT License. See the LICENSE file for details.

References
-----------
For detailed descriptions of algorithms and methodologies used in this implementation, please refer to the publications associated with this codebase.
 This README provides a comprehensive overview of your CNetSOM project, detailing its functionality, usage instructions, and structure.