BME Seminar Series: Bo Ning
Content
Engineering Proteins to See Diseases: From detecting with the molecules nature gave us to generating the ones we need for diagnostics
Bo Ning, PhD
Tulane University
Thursday, September 17
Morrill III rm. 203
1:00-2:15 PM
Every molecular diagnostic rests on one component: a recognition protein that binds the target and hands off to a readout. For the past decade, the Ning Lab has built diagnostics around the recognition molecules nature provides, antibodies, nanobodies, and CRISPR nucleases, engineering the assay chemistry around each one's strengths and limits. That work has produced a CRISPR-based blood test for tuberculosis that earned FDA Breakthrough Device designation, extracellular-vesicle liquid biopsies for cancer and infectious disease, and on-chip assays that read antigen-specific T-cell activation directly from patient samples.
But natural recognition molecules have hard ceilings: some targets never raise a usable antibody, Cas nucleases are constrained by a fixed PAM and inherited activity, and discovery by immunization or screening is slow. This talk describes a shift now underway in the lab, from discovering recognition proteins to generating them. I will introduce two complementary generative approaches: protein language models, which learn the sequence grammar of a protein family and sample novel members of it, and structure-based design, which builds a binder from a target's geometry outward using RFdiffusion, ProteinMPNN, and AlphaFold2.
Two ongoing student projects illustrate where this stands. The first fine-tunes a protein language model on the CRISPR-Cas Atlas to generate engineered Cas12a nucleases, with matching guide-RNA scaffolds for next-generation CRISPR diagnostics. The second uses structure-based design to build a de novo binder for a small peptide toxin, a target class where antibodies typically fail. Both projects are candid works in progress, and I will walk through what is working, where each currently stalls, and what the results imply for a diagnostics platform in which the recognition molecule is no longer a fixed input, but a design variable.
About Bo Ning: Dr. Bo Ning is an Assistant Professor in the Department of Biochemistry & Molecular Biology and Co-Director of the Center of Cellular and Molecular Diagnosis at Tulane University School of Medicine. His lab develops molecular diagnostics that detect disease earlier and more affordably, working across CRISPR-based nucleic acid detection, extracellular-vesicle liquid biopsies, and on-chip immune-response assays. His work on a CRISPR-based blood test for tuberculosis received FDA Breakthrough Device designation in 2024. His research has been published in journals including Nature Nanotechnology, Nature Biomedical Engineering, Nature Protocols, Science Translational Medicine, Science Advances, Am. J. Respir. Crit. Care Med. and The Lancet Microbe, and is supported by the NIH and an institutional COBRE center grant. His lab's current work extends this line of research toward generative protein design, using both protein language models and structure-based design methods to create new recognition proteins for diagnostic applications.