2DCDRCNN: Cancer Drug Response Prediction using CNN and Gene Expression Data

2DCDRCNN is a deep learning model that utilizes Convolutional Neural Networks (CNNs) to predict the IC50 value of a specific drug on a given cancer cell line. This model takes advantage of one-hot encoded drug data and t-SNE transformed gene expression data to make accurate predictions about the effectiveness of drug-cancer interactions. By learning from known interactions between drugs and cancer cell lines, the project aims to enhance drug development and tailor treatments for individual patients.

Table of Contents

• Introduction
• Features
• How It Works
• Installation
• Usage
• Example
• Contributing
• License

Introduction

One element that contributes to the complexity and challenges of cancer treatment is the variation in patient responses based on their unique genetic and epigenetic make up. The 2DCDRCNN project addresses this issue by leveraging molecular structures of drugs and gene expression profiles of cancer cell lines to predict the effectiveness of new drug-cancer combinations. This predictive model can help researchers and clinicians identify potential candidates for testing and tailor treatments for individual patients.

Features

• Predict Drug-Cancer Interaction: Predict the IC50 value of a specific drug on a given cancer cell line.
• Utilize Gene Expression Data: Utilize t-SNE transformed gene expression data for accurate predictions.
• Enhance Drug Development: Test approved drugs and candidates for their estimated effectiveness on different cancer cell lines.
• Personalized Treatment: Predict patient responses to various cancer drugs using gene expression profiles.

How It Works

1. Data Preparation: The model requires one-hot encoded drug data and t-SNE transformed gene expression data as input.

2. Convolutional Neural Network (CNN): The CNN architecture is used to learn patterns and relationships between drugs and gene expression profiles.

3. Prediction: Given a drug and gene expression data, the trained model predicts the IC50 value of the drug on the specific cancer cell line.

4. Enhanced Drug Testing: The model's predictions can be used to identify potential drug candidates for testing and streamline drug development processes.

Installation

1. Clone the repository to your local machine:

   git clone https://github.com/yourusername/2DCDRCNN.git
   cd 2DCDRCNN

2. Set up your development environment with the required dependencies.

Usage

1. Prepare your one-hot encoded drug data and t-SNE transformed gene expression data. Mount to your respective google drive

2. Run the model using your preferred Python interpreter:

   python Model.py

Example

Predicting IC50 value for a drug-cancer combination:

Enter one-hot encoded drug data: [0, 1, 0, ...]
Enter t-SNE transformed gene expression data: [0.123, 0.456, ...]

Predicting IC50 value...
Estimated IC50 value: 5.67

Contributing

Contributions are welcome! If you'd like to contribute to this project, please follow the guidelines in CONTRIBUTING.md.

License

This project is licensed under the MIT License.

---

For any questions or suggestions, please contact [your name/email here]. Happy predicting!