🧠 Deep Transfer Learning for Molecular Cancer Classification

This repository provides a MATLAB implementation of a deep transfer learning framework for molecular cancer classification using gene expression data. The approach leverages deep neural networks (DNNs) and transfer learning to improve classification accuracy across various cancer types, particularly when training data is limited.

---

📄 Reference Publication

If you use this code, please cite the following paper:

Sevakula, R. K., Singh, V., Verma, N. K., Kumar, C., & Cui, Y. (2018)
Transfer learning for molecular cancer classification using deep neural networks
IEEE/ACM Transactions on Computational Biology and Bioinformatics, 16(6), 2089–2100.
📄 DOI: 10.1109/TCBB.2018.2816014

@article{sevakula2018transfer,
  title={Transfer learning for molecular cancer classification using deep neural networks},
  author={Sevakula, Ram K and Singh, Vikas and Verma, Nishchal K and Kumar, Chetan and Cui, Yidong},
  journal={IEEE/ACM Transactions on Computational Biology and Bioinformatics},
  volume={16},
  number={6},
  pages={2089--2100},
  year={2018},
  publisher={IEEE}
}

🚀 Features
Deep neural network model for cancer type classification

Transfer learning strategy for small and imbalanced datasets

Works on microarray or RNA-seq gene expression datasets

Fully implemented in MATLAB

🛠️ Installation & Usage
📥 Clone the repository

git clone https://github.com/vikkyak/Deep_Transfer_Learning.git

▶️ Run in MATLAB
Open MATLAB

Navigate to the Deep_Transfer_Learning folder

Run the main script:

run Proposed_Classification_Procedure.m
Modify main.m or related scripts to load your own dataset.

📁 Directory Structure
File/Folder	Description
main.m	Main script to run deep transfer learning model
functions/	Custom functions for preprocessing and modeling
data/	(Optional) Sample gene expression data
models/	Saved model weights and architectures (if any)

📧 Contact
Vikas Singh
📧 vikkysingh07@gmail.com