AI for Polarizability-Based Biosensors

Inspiration

Polarization-based biosensors have achieved single-molecule sensitivity, but in the process, the specificity of the sensor signal data has been compromised. This project aims to decode the sensor signal data by using predictive models rooted in physics.

What it does

The new project tmm-sensors on GitHub couples classical and quantum approaches through the polarizability tensor of single molecules to model the far-field radiation patterns of single molecules above complex materials. The polarizability tensor for small organic molecules (up to 23 atoms in size) can be obtained from the QM7-X database.

What's next for AI for Polarizability-Based Biosensors

To improve the range of molecules that can be studied using tmm-sensors, I propose to explore the correlation between molecular structure and polarizability with machine learning models and extrapolate the QM7-X database to simulate larger organic molecules, whilst preserving quantum simulation accuracy.

Built With

• python
• qm7-x