Paul P. Chevis Endowment Seminar Series
Engineering Proteins to Improve Human Health and Protect Crops
Daniel Woldring
Chemical Engineering and Materials Science
Michigan State University
Abstract
Engineering useful proteins requires navigating vast sequence spaces while learning from limited experimental data. The Woldring laboratory develops integrated experimental and computational approaches that connect protein sequence, structure, and biological function across medicine and agriculture. This presentation describes how molecular representations, evolutionary information, and experimental validation enable predictive and generative protein engineering.
Antibody repertoire analysis illustrates the clinical relevance of sequence representations. Biochemical motifs distinguish severe pediatric traumatic brain injury samples from controls in a small cross-validation cohort, motivating further biomarker validation. Complementary studies establish how protein encodings, sampling strategies, and ensemble learning influence fitness prediction. For molecular interactions, BINDsmart interface representations and HOLIgraph integrate ligand chemistry with protein contacts to improve the prediction of membrane transport protein inhibition. Evaluations of model robustness and reproducibility identify limitations that must be addressed before computational predictions can reliably guide experiments.
Building on these principles, EvoSeq combines ancestral sequence reconstruction with generative modeling to produce enzymes with increased thermostability and retained, although reduced, catalytic activity. Plant immunity studies reveal how glycosylation constrains FLS2 expression and ligand recognition, while structural features help predict peptide signaling outcomes. A design strategy transfers structural knowledge from a natural inhibitor complex to engineer protein candidates that reduce fungal spore germination.
Together, these studies demonstrate how experimentally grounded modeling can progress from interpreting biological information to creating proteins with useful properties. Future work will strengthen iterative design and validation, including engineering membrane transporters for selective molecular uptake and eventual tissue delivery, while translating research capabilities into interdisciplinary training and biochemical engineering education at MSU.
Bio
Prof. Daniel Woldring is an Assistant Professor in ChEMS at MSU. He grew up in West Michigan and earned his B.S. in Chemical Engineering with a minor in Mathematics from Michigan Technological University. He then completed his Ph.D. in Chemical Engineering at the University of Minnesota in Minneapolis before pursuing postdoctoral training in structural biology with Dorothy Kern at Brandeis University as an HHMI Fellow. He joined the faculty at Michigan State University in 2019. Dr. Woldring’s research combines protein engineering, structural biology, machine learning, and molecular simulation to design proteins with new or improved functions for applications in human health, agriculture, and biotechnology. His group has received funding from the National Institutes of Health to engineer antibodies for neuroblastoma, from the USDA and Michigan agricultural commodity groups to develop proteins that protect crops from bacterial and fungal pathogens, and from DARPA to develop machine learning approaches for predicting protein function based on dynamics. His teaching interests include biochemical engineering, protein engineering, and the integration of machine learning and computational tools into engineering education. He particularly enjoys mentoring students interested in working at the intersection of engineering, biology, and data science.
Zoom: https://msu.zoom.us/j/96728843280
Passcode: Seminar
Persons with disabilities have the right to request and receive reasonable accommodation. Please call the Department of Chemical Engineering and Materials Science at (517) 355-5135 at least three days prior to the seminar; requests received after this date will be met when possible.