Thema der Dissertation:
Graph-Theory Algorithms for Dynamic Hydrogen-Bonded Networks in Proteins and Lipid Membranes
Graph-Theory Algorithms for Dynamic Hydrogen-Bonded Networks in Proteins and Lipid Membranes
Abstract: Computer simulations can give essential insights into the dynamics of biomolecular systems but raise significant big-data challenges. To overcome the challenge of large data sets combined with the complexity of biomolecular interactions, I implemented a set of robust algorithms inspired by graph theory that allows us to use large data sets from atomistic molecular dynamics (MD) simulations and derive simple graphical representations of the hydrogen bond (H-bond) networks of lipid membrane models, proteins in different intermediate states, and of the response of the proteins to mutations. Our algorithms facilitate highly efficient analyses of dynamic H-bond networks at the lipid membrane interface. We introduce the implementation of a Connected Components algorithm to cluster lipid molecules and a Depth First Search (DFS) algorithm that allows us to characterize the topology of dynamic H-bond clusters sampled by lipid headgroups in MD simulations. Our algorithms are further extended to study conformational dynamics in proteins. An example is SecA, a protein motor that couples Adenosine triphosphate (ATP) binding and hydrolysis with the pre-protein substrate's translocation through the membrane embedded SecYEG protein translocon. We present a methodology of applying graph-based approaches to characterize the dynamics of the SecA protein motor by computing long-distance H-bond pathways that inter-connect the nucleotide-binding pocket and the pre-protein binding site, shortest-distance routes and centrality measures that reveal amino acids with a central role in the total connectivity of the protein graph. Our algorithms are applied to the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-COV-2) protein S crystal structures. Protein S undergoes conformational changes and symmetry loss of core H-bonded clusters as it transitions from the closed to the pre-fusion conformation.
Time & Location
Mar 04, 2024 | 01:00 PM
Hörsaal A (1.3.14)
Fachbereich Physik, Arnimallee 14, 14195 Berlin