AI in Antibody Discovery and Development

ABSTRACT

Therapeutic antibodies remain a cornerstone of modern medicine, transforming the treatment landscape across oncology, immunology, and infectious diseases. However, traditional antibody discovery and development workflows remain largely empirical, iterative, and resource-intensive. Candidates that demonstrate promising target binding often fail later due to biophysical and pharmacological liabilities—such as aggregation, high viscosity at formulation-relevant concentrations, poor stability, or suboptimal bioavailability—leading to increased cost, extended timelines, and attrition during translation to the clinic. In this seminar, I will discuss how artificial intelligence is reshaping this paradigm by enabling predictive, data-driven decision-making throughout the antibody development pipeline. By integrating machine learning, deep learning, and sequence-based modeling with fundamental biophysical principles, AI models can extract meaningful patterns from large experimental datasets and link molecular features to functional outcomes. Modern protein language models and structure-aware neural networks now allow us to move beyond descriptive correlations toward scalable prediction of structure, interaction propensity, and developability risk directly from sequence.

BIOGRAPHY

Dr. Pin-Kuang Lai is an Assistant Professor in the Department of Chemical Engineering and Materials Science at Stevens Institute of Technology. He received his B.S. and M.S. degrees in Chemical Engineering from National Taiwan University in 2011 and 2013, respectively, and completed his Ph.D. in Chemical Engineering at the University of Minnesota in 2018. His doctoral research focused on developing computational and experimental approaches to study antimicrobial peptides. From 2018 to 2021, Dr. Lai conducted postdoctoral research at MIT, where he investigated antibody stability at high concentrations.

Dr. Lai has collaborated with leading pharmaceutical companies, including AstraZeneca, Merck, Sanofi, Takeda and Johnson & Johnson, as well as national laboratories such as NIST and Brookhaven. His work focuses on advancing computational tools, including machine learning and molecular simulations, alongside biophysical characterization methods such as NMR and SAXS, to predict and analyze antibody aggregation and viscosity at high concentrations. Dr. Lai was named 2024 Top Scholar by ScholarGPS in Antibody Research. Dr. Lai was a Keynote Speaker at PEGS Boston, Bioprocessing Summit and AAPS National Biotechnology Conference. At Stevens, Dr. Lai contributes to AI-focused department-/institute-wide services such as working for the Stevens Institute for Artificial Intelligence and serving in the Working Group on Core AI Grad Curriculum. In addition to his academic and research endeavors, Dr. Lai serves as an Assistant Editor for mAbs, a leading journal in antibody research and development.