Computational biology for natural products and proteins

Highly-accurate virtual screening with AI

We are using machine-learning techniques in addition to the traditional virtual screening methods (e.g. molecular docking, pharmacophore modeling, ligand-based shape matching, etc) to develop an integrated and high-accuracy virtual screening method.

We are participating in CRITICAL ASSESSMENT OF COMPUTATIONAL HIT-FINDING EXPERIMENTS (CACHE) challenge (https://cache-challenge.org/)

Computational protein design and modeling

Natural proteins evolved over millions of years to exert appropriate functions maintaining biological system but, at the same time, highly-limited to that. Nature doesn't care unnecessary protein interactions and doesn't care protein stability at 100 C. We engineer super stable and functional proteins by combining computational protein design (Rosetta - https://www.rosettacommons.org/) with experimental directed evolutionary approach (e.g. phage display).

Microbiome interaction analysis

Microbiome consists of complicated interactions between numerous microbes. We are developing new computational tools to decipher hidden functional microbe interactions.