NeuroBoostedKcat: Predicting Transferase Kcat Values

Overview

Model stacking neural network + lightgbm model to predict transferase kcat values.

Project Description

In the dynamic and evolving realm of enzyme kinetics, quantifying the turnover number (Kcat) for Transferases plays an integral role in understanding their catalytic efficiency. NeuroBoostedKcat represents an avant-garde approach, converging deep learning with traditional boosting methodologies to predict Kcat values with enhanced precision.

Technologies Used

• Python
• TensorFlow/Keras
• LightGBM
• XGBoost
• Neural Network

---

Table of Contents

1. Installation
2. Data Preparation
3. Model Architecture
4. Usage
5. Results
6. Contributing
7. License

---

Installation

git clone https://github.com/your_username/NeuroBoostedKcat.git

cd NeuroBoostedKcat

pip install -r requirements.txt

---

Data Preparation

Explain how the dataset should be prepared. ProTrans

---

Model Architecture

Sequential Neural Network

Describe the architecture of your Sequential Neural Network, its layers, and their functionalities.

LightGBM

Explain the parameters and structure of your LightGBM model.

XGBRegressor

Detail the final stacking model using XGBRegressor.

---

Usage

Provide examples of how to use the model for predicting Kcat values.

```python from NeuroBoostedKcat import predict_kcat

Example usage

predict_kcat(sequence='your_sequence_here') ```

---

Results

Coming soon

Contributing

Lok Weng Bryan Cheong Lok.weng.b.cheong@vanderbilt.edu

---

License

"""