🚀 Hack-The-Pharma!

Our team, Hack-The-Pharma, developed a data-driven solution to Challenge 1: Predict Kinase Selectivity Using Machine Learning for PharmaHacks 2025. Challenge Repository. Our project aims to address critical challenges in pharmaceutical data analysis and molecular property prediction, using Python and R. In drug development 💊, predicting molecular properties is crucial for screening potential drug candidates efficiently. Our solution leverages machine learning techniques to forecast kinase-inhibitor interactions from molecular properties and chemical structures.

---

📦 Installation and Setup

1. Clone the Repository

git clone https://github.com/Swagat404/Hack-The-Pharma.git
cd Hack-The-Pharma

2. Create a Virtual Environment (Optional)

python -m venv venv
source venv/bin/activate   # On macOS/Linux
.\venv\Scripts\activate    # On Windows

3. Install Python Dependencies

pip install -r requirements.txt

4. Install R Packages From your R console:

install.packages(c("dplyr", "ggplot2", "caret", "randomForest"))

---

▶️ Running the Project

To run the main pipeline:

python main.py

To execute R scripts:

Rscript script_name.R

🚧 Future Improvements

• Enhanced feature selection for better model accuracy
• Integration with external chemical databases to enrich available input data
• Better script organization for testing calls