Investigators and Program Directors

Research Interests

Biological systems exist through complex interplay of a diverse range of macromolecules performing specific functions. For each macromolecule, the gap between visualizing its atomic-level three dimensional structure and understanding its functional properties needs to be traversed by a description of the conformational free energy landscape governing its activity. The assembly and function of large macromolecular complexes relies on molecular recognition of individual macromolecules by their partners and dynamic control of their interaction with substrates. The primary goal of our research is to use computational calculations and refined analysis techniques to optimally extract free energy landscapes describing biologically relevant macromolecular conformational change. We also develop techniques to facilitate validation of computational predictions with structural and biochemical data. Our research is focused on the following areas:

Nucleic acid base recognition and protein interaction:

Base pairing, or its reverse (base pair separation), is the main molecular mechanism involved in the three primary processes in the central dogma of biology: replication, transcription and translation. Even a small error in base recognition can result in drastic malfunctions in all downstream processes. The sequence context of each base can have a large effect on its recognition by its pairing partner. The specific protein environment also plays an important role in facilitating the dynamic changes required in base recognition. We aim to establish a quantitative and atomistic understanding of these dynamic and molecular recognition processes involved in nucleic acid function. The methods developed are applied: (a) to better characterize the recognition of specific features of small interfering RNA (siRNA) by proteins involved in RNA interference (RNAi); (b) to delineate the complex dynamic mechanism by which single base deletion or addition mutations occur in DNA strand extension by Y-family DNA polymerases (in collaboration with Dr. Janice Pata)

Chemistry and dynamics of intein splicing:

Inteins are intervening protein sequences that interrupt other protein sequences but can splice themselves out from their surrounding protein segments (called exteins). Over 450 intein occurrences have been identified in multiple pathogenic and non-pathogenic microorganisms, including Mycobacterium tuberculosis, suggesting inteins as potential therapeutic targets. Their unique self-splicing ability also makes them powerful and versatile protein engineering tools. Three classes of catalytic mechanisms have been proposed for the unique protein rearrangements involved in intein self-splicing. The precise structural and dynamic details of these splicing mechanisms and their modulation by mutations in proximal/distal extein or intein regions is not yet well-understood. Our work (in collaboration with Dr. Marlene Belfort and Dr. Patrick Van Roey, along with other scientists in the Wadsworth-RPI Intein Consortium) is to characterize and experimentally validate the chemical and dynamic details of intein self-splicing mechanisms. This knowledge is then to be applied for identifying new transition state analog scaffolds for intein inhibitors, and for creating a heuristic model based on implicit solvent calculations to predict intein splicing activity.

Nucleotide-dependent protein conformational change:

A ubiquitous biological mechanism involves protein conformational changes caused by binding and cleavage of the terminal phosphate of specific nucleotides (e.g. ATP, GTP). We aim to probe the details of this specific mechanism by determination of effects of presence or absence of this terminal phosphate on the free energy landscapes of the nucleotides and nucleotide-binding proteins in their various ligand-bound conformational combinations.

Hybrid-approach structure determination methods:

In spite of continuing advances, structure determination of large macromolecular assemblies is still an extremely challenging problem for conventional techniques such as X-Ray crystallography or Nuclear Magnetic Resonance spectroscopy. Combination of these techniques with lower resolution techniques such as cryo-electron microscopy or cryo-electron tomography or proteomics techniques such as mass spectrometry can address this problem across a broad spectrum of length scales. There is a critical need for multiple computational strategies that provide a central quantitative protocol to combine data from disparate experimental techniques to obtain probability-weighted high-resolution models. We aim to develop and optimize such computational strategies by pursuing accurate probability-weighted structure determination of large macromolecular assemblies such as the dynein motor protein (in collaboration with Dr. Michael Koonce) and the mammalian mitochondrial ribosome (in collaboration with Dr. Rajendra Agrawal, Dr. Linda Spremulli (University of North Carolina-Chapel Hill), Dr. Joachim Jaeger, and Dr. Patrick Van Roey).

Our long range goals are to incorporate detailed atomic-scale understanding of conformational change into simplified coarse-grained models to scrutinize how macromolecular components interact and coordinate their activity to form functional nanoscale cellular machinery.

Contact Information

Telephone: (518) 474-0569
Fax: (518) 402-4623
E-mail: banavali@wadsworth.org