Richard Shuai

I'm a first year Biophysics PhD student at Stanford studying machine learning applications in protein engineering. Previously, I studied Computer Science and Statistics at UC Berkeley and worked under Professor Nilah Ioannidis to study the effects of noncoding variants on chromatin accessibility using machine learning.

In Professor Jeffrey Gray's lab at Johns Hopkins, I've developed deep learning approaches for designing antibody sequences under the mentorship of PhD student Jeffrey Ruffolo. We aim to generate synthetic antibody libraries that can accelerate the development of therapeutic antibodies.

Email  /  CV  /  Google Scholar  /  Twitter  /  Github

Using a deep generative language model trained on ~500 million antibody sequences to design synthetic antibody libraries.

Used deep learning for real-time image reconstruction from convolutions with a spatially-variant point spread based on a U-Net with learnable Wiener deconvolutions. Project page.

Early work on using deep learning for real-time image reconstruction from convolutions with a spatially-variant point spread.

Template taken from Jon Barron's page (source code).