Accelerating Drug Discovery with Deep Learning

We develop software to perform common drug discovery tasks while leveraging deep learning to achieve state-of-the-art results. These include gnina, which performs molecular docking (and relevant subroutines), and libmolgrid, a C++ library with Python bindings that facilitates machine learning with molecular data (especially when using multidimensional grids, like voxels, for molecular representation).

Next

Accelerating Drug Discovery Through Algorithmic Innovation

We develop novel methods that perform drug discovery tasks more efficiently. This includes pharmit, which performs a pharmacophore search through a database that scales with the query complexity rather than database size. We develop and utilize parallel algorithms to accelerate all projects, including the voxelization and grid processing routines that are critical to gnina.

Next

About me

I am a PhD candidate in the Joint Carnegie Mellon/University of Pittsburgh PhD program in Computational Biology. I work with David Koes. My undergraduate degrees are in physics and English. In the past, I did research with Brian D'Urso. As part of that project, I programmed a microcontroller to function as a lock-in amplifier, signal generator, and oscilloscope. I previously drafted employment-based immigration petitions at May Law Group, worked with elementary students through Allegheny Children's Initiative, was a Crew Leader for Pittsburgh Summer Youth Employment Program via the SCA, and worked as an organist throughout the Altoona-Johnstown Diocese, among other things.

Next