Tongqing Zhou, Ph.D.

The Structural Bioinformatics Core (SBIC), headed by Dr. Reda Rawi, employs state-of-the-art computational techniques, spanning from sophisticated structural modeling to generative artificial intelligence (AI) and machine learning (ML) to create next-generation vaccine immunogens, therapeutics, and probes. Our collaborative, interdisciplinary approach ensures that computational insights seamlessly feed into laboratory workflows, expediting the discovery and design process while reducing costs. Focused on critical public health concerns like HIV-1 and influenza, and remaining agile to counter emerging pathogens, SBIC plays a pivotal role in advancing targeted, effective innovative solutions that align with NIH’s mission to protect and improve global health.

Structural Bioinformatics and Therapeutic Design

Artificial Intelligence-Based Protein Design

We employ advanced AI/ML frameworks to develop innovative antibodies, immunogens, and probes with therapeutic applications. To enhance these designs, we apply molecular dynamics (MD) simulations, which allow for the exploration of conformational flexibility and offer valuable insights into protein-protein interactions. By continuously incorporating experimental data, we effectively reduce both time and costs while enhancing immunogenicity, specificity, and overall stability. This integration of AI, MD, and laboratory validation positions SBIC at the leading edge of creating next-generation therapies for HIV-1, influenza, and other emerging viral threats.

Structural Bioinformatics

Our suite of computational methods elucidates the atomic details governing protein function and stability. We utilize molecular modeling and MD simulations to capture conformational changes, reveal transient states, and characterize binding interfaces essential for therapeutic efficacy. Techniques like steered and targeted MD enable us to investigate key conformational transitions in viral proteins, including HIV-1, offering valuable insights into how structural shifts affect antibody binding and neutralization. We also develop in silico tools to map glycan density on glycoproteins, correlating glycosylation patterns with properties like immunogenicity, antigenicity, and effective recognition distance. These combined approaches guide the rational design of therapeutics and diagnostics, ensuring that computational discoveries lead to impactful experimental outcomes.

Genomics

Beyond protein-level insights, SBIC extends its focus to the genomic underpinnings of immunity and pathogenicity. We wield next-generation sequencing to identify viral and host mutations that drive immune escape and uncover potential targets for vaccine and therapeutic intervention. This helps us track viral evolution, detect emerging variants of concern, and prioritize antigens most likely to elicit protective immune responses. Our close collaboration with experimentalists ensures that computational predictions are validated in the lab, ultimately contributing to advancements against global health threats.

Biological Data Science

As data grow in volume and complexity, SBIC applies mathematical, statistical, and machine learning techniques to garner actionable insights. We advise on study design, execute power analyses to calculate necessary sample sizes, perform robust statistical analyses, and develop predictive models—using linear and polynomial regression, tree-based methods, and support vector machines. We leverage protein language models for feature engineering to capture sequence and structural nuances, apply feature selection techniques for marker discovery, and use high-dimensional statistical analyses. These data-driven strategies maintain a scalable research pipeline that bolsters NIH’s leadership in biomedical innovation.