Research Interests
  • Development and application of NMR methods for studying the structure, dynamics, and function of proteins and small molecules [projects]
    Proteins do not act in isolation. Rather, they participate in interaction networks with other proteins, DNA, RNA, membranes and small ligands, which dictates their biological function. These interactions and binding events typically have both energetic and entropic consequences which result in a change of protein structure and its associated protein dynamics. NMR spectroscopy is extremely powerful at measuring the structural dynamics of proteins on a large range of time scales. Our lab uses many different NMR methods together with other biophysical tools to study protein behavior under different conditions and in the presence and absence of binding partners and ligands. We are particularly interested in the high resolution spatial and temporal description of dynamic processes, from picoseconds to milliseconds and beyond. Such information is provided by NMR spin relaxation parameters that can be measured for almost every atom in a protein and residual dipolar couplings (RDCs) that provide unique structural and dynamic information on the slower times scales.
    To understand the NMR measurements on a fully quantitative level, we are developing and applying dynamics models, some are analytical, such as the 3D GAF model and local and collective contact models, while others are computational using molecular dynamics simulations together with appropriate post-processing methods such as Reorientational Eigenmode Dynamics (RED) and methods to estimate related thermodynamic quantities in particular conformational entropies (reorientational and 2D entropy methods).
  • Fast NMR methods (covariance NMR) [projects]
    Multi-dimensional NMR spectroscopy provides resolution enhancement by spreading NMR resonances into additional dimensions, which decreases the chances of cross peak overlaps. However, the acquisition of each additional dimension is expensive in terms of NMR spectrometer time. We have introduced covariance NMR spectroscopy, which is a new framework for the representation of two-dimensional NMR data by representing them in the form of a covariance matrix. Due to its symmetry, the covariance spectrum has intrinsically high resolution along the indirect dimension requiring minimal measurement time. A benefit of the method is that it allows direct comparison with the corresponding Fourier transform spectrum. Covariance spectroscopy, which is continuously further developed and generalized, now includes indirect covariance spectroscopy, double-quantum covariance spectroscopy, regularized covariance spectroscopy, and covariance spectroscopy in four dimensions.
  • Quantitative and automated analysis of metabolomic mixtures [project]
    Metabolomics is a newly emerging field at the interface of chemistry, biology, and the computational sciences. It represents the study of the interactions between the chemical components of a biological system, such as a cell, tissue, or organism, with the goal to understand how these interactions give rise to the behavior of the system as a whole.
    These systems constitute complex chemical mixtures whose fingerprinting in terms of identification of chemical components together with their concentrations as a function of stress, age, and disease is a primary objective. To relate this information to the precise biological state of the system, new methods are needed for the accurate, efficient, and robust profiling of correlated and anti-correlated changes of distinct sets of biomarkers. Translational research in this area will lead to personalized medicine that delivers effective therapies tailored to the precise biological state of an individual. Nuclear magnetic resonance (NMR) spectroscopy has a unique and vast potential for both fingerprinting and profiling because of its highly quantitative nature and because it does not require physical separation of the components. Covariance NMR spectroscopy shows excellent potential for the quantitative and fully automated analysis of biofluidic mixtures. The combination of covariance total correlation spectroscopy (covariance TOCSY) with a novel spectral deconvolution method, termed ‘DemixC’, provides a set of unique traces that in good approximation correspond to the 1D spectra of the individual components of the mixture. Importantly, this analysis procedure does not require physical separation of the components. The high sensitivity and high resolution afforded by proton covariance NMR along with the analysis algorithm makes the method ideally suited for fully automated high-throughput applications in metabolomics. The method can be readily applied to the analysis of complex mixtures encountered in living systems including cell cultures and mice disease models, the environment, and food to study metabolic fingerprints and their variations upon changing conditions, such as age, stress, obesity, disease, and chemical and drug treatment.

    In all these applications the group entertains close contact with the National High Magnetic Field Laboratory where many of the measurements are carried out and where advances in hardware and software play an important role for the success of these projects. The group is actively involved in collaborations with scientists across the US, Europe, and Asia.

Featured Research
Molecular dynamics by nuclear magnetic resonance (NMR)
Covariance NMR spectroscopy Quantitative analysis in metabolomics
This research is funded by the National Science Foundation and the National Institutes of Health.