Programming Resources

Relevant Resources Based on Published Material:

Algorithms: As the rise of computational advances in artificial intelligence (AI) and machine learning (ML) are rapidly integrating into chemistry and nanotechnology, we believe that providing transparency in programming can provide both rigor and reproducibility in research. Therefore, we are sharing our algorithms that we make in-lab that are used in our research. If you use our code please cite the respective reference literature for source code.

• Machine Learning Powered Nanosensor Array, MILAN (AGONS): PyPi, Github, Reference Literature. (Active Project)

  • We developed MILAN as a combination of ML and our biomarker-free 2D-DNA nanotechnology sensor array to automate and improve the predictive/classification capabilities of various biological materials. In doing so, we created our algorithmically guided nanosensor selector (AGONS) through Python packages facilitate the machine learning process. Overall, we found that MILAN can predict from 90 - 100% classification accuracy while reducing the size of a nanosensor array from multiple pure and real sample datasets. We believe that the powerful utility of MILAN is not solely limited to 2D nanoparticles but can be expanded towards the development and design of other nanosensor arrays.