Postdoctoral research fellow
National Institutes of Health
I am currently a postdoctoral research fellow in the Drug Metabolism and Pharmacokinetics Core within NCATS’ Division of Preclinical Innovation, where my primary focus is on the use of machine learning methods with in vitro absorption, distribution, metabolism and excretion (ADME) datasets for developing in silico ADME models to predict novel compound properties. By applying in silico ADME models in designing drug-like molecules and identifying lead candidates, I aims to accelerate the discovery of new drugs for unmet medical needs.
I earned Ph.D. in bioinformatics from the University of Science and Technology (UST) in South Korea. During Ph.D., I focused on developing methods for predicting new or active binding sites for ligand utilizing big-data of 3D protein structures.
Following my graduate studies, I continued research as a postdoctoral researcher at the Korea Research Institute of Chemical Technology (KRICT), where I focused on the repositioning and repurposing of existing compounds for new indications and targets through the analysis of gene expression data. Based on my research in protein-compound interactions, I expanded the scope of my research by applying machine learning techniques to various biological data sources to develop more accurate prediction models.
My research primarily focuses on accurately predicting the efficacy of novel drug candidates before they advance to clinical trials, as well as helping researchers identify new drug candidates more quickly and reduce the time and costs associated with the drug discovery process. To achieve these goals, I leverages a range of data sources to screen drug compounds that are safe and effective for human use by considering factors such as absorption, distribution, metabolism, excretion and toxicity (ADMET) with state-of-the-art machine learning methods. I also focuses on identifying compounds that can efficiently bind to the active site of proteins, using protein structure analysis as a key tool.