Ho-Joon Lee, PhD

My current research is centered around biomedical data science and single cell biology, focusing on machine learning methods. My general interests lie in biomedical artificial intelligence and fundamental biology using tools from diverse fields including deep learning, systems/network biology, and physics.

Three major on-going projects are (1) machine/deep learning of electronic health records and MRI images for ischemic stroke etiology classification with Dr. Richa Sharma of neurology (1 patent pending; https://www.researchsquare.com/article/rs-3367169/v1) (2) single-cell systems immunology of West Nile virus infection with the Montgomery and Kleinstein labs (https://www.biorxiv.org/conten...) (3) single-cell multi-omics data analysis of zebrafish brains and embryos with the Giraldez lab (https://elifesciences.org/arti...).

In response to the COVID-19 pandemic, I have been working on virus-host protein-protein interactions (PPIs) and early drug discovery in silico together with Dr. Prashant Emani (the Gerstein lab) as a COVID HASTE working group of the Yale School of Engineering and Applied Science (see Figure 1 below for an overview; 1 patent pending). The project on virus-host PPIs was partially supported by a seed fund from the Northeast Big Data Innovation Hub and a preprint has been published (https://nebigdatahub.org/a-lan...). A preprint on ligand-protein binding affinity prediction is also available (https://arxiv.org/abs/2310.03946). As an additional effort, with help from two of my former colleagues, Dr. Vinayagam Arunachalam (Pfizer Inc.) and Dr. Yang-Yu Liu (Harvard Medical School, Brigham and Women's Hospital), I carried out controllability analysis of a directed human protein-protein interaction network for SARS-CoV-2 based on our previous paper (https://www.pnas.org/doi/10.10...). A preprint is available here, https://www.biorxiv.org/conten....

My previous research mostly concerned biological questions in systems biology and network medicine by developing algorithms and models through a combination of statistical/machine learning, information theory, and network theory applied to high-throughput multi-dimensional data. It covered genomics, transcriptomics, proteomics, and metabolomics from yeast to mouse to human for integrative analysis of regulatory networks on multiple molecular levels. I previously carried out proteomics and metabolomics along with a computational derivation of dynamic protein complexes for IL-3 activation and cell cycle in murine pro-B cells (https://www.cell.com/cell-repo...; see Figure 2 below), for which I developed integrative analytical tools using diverse approaches from machine learning and network theory. My ongoing interests in methodology include machine/deep learning and topological Kolmogorov-Sinai entropy-based network theory, which are applied to (1) multi-level dynamic regulatory networks in immune response, cell cycle, cancer metabolism, and cell fate decision and (2) single cell-based and mass spectrometry-based omics data analysis. Two specific projects were (1) Dynamic metabolic network modeling of a mammalian cell cycle using multi-omics time-course data in collaboration with the Chandrasekaran lab at the University of Michigan (see Figure 3 below; https://www.biorxiv.org/conten...; https://www.cell.com/iscience/...) and (2) Tri-omics analysis of macrophage polarization in pancreatic cancer in collaboration with the Lyssiotis lab at the University of Michigan Medical School (https://elifesciences.org/arti...).