Inspiration

Drug discovery is costly and time-consuming. We wanted to harness AI to streamline the process by leveraging molecular databases and predictive models to accelerate bioactive molecule identification.

What it does

BioMimic predicts the biological activity of molecules using AI models trained on databases like PubChem and ChEMBL, providing insights that help researchers identify potential drug candidates.

How we built it

We used Python, TensorFlow, and PyTorch for AI modeling, Streamlit for the web interface, and integrated molecular databases to provide a seamless user experience.

Challenges we ran into

Handling large datasets, optimizing ML models for accuracy, and integrating multiple data sources into an efficient workflow were key challenges we tackled.

Accomplishments that we're proud of

Successfully building an AI-driven platform that simplifies bioactive molecule discovery and deploying an intuitive web interface for easy access.

What we learned

We gained insights into optimizing ML models, managing large bioinformatics datasets, and enhancing user experience through iterative design and testing.

What's next for BioMimic

We aim to expand the dataset coverage, improve prediction accuracy, and collaborate with research institutions to bring BioMimic to real-world applications.

Built With

Share this project:

Updates