Academic research

Protein Structure and Dynamics

Proteins play central roles in many significant biological processes in living organisms from cell signaling to immune response. Being in the center of life, proteins have been a focus of scientific research since they were first proposed in mid-19th century. Scientists now have a good understanding of how proteins differ from one another (amino acid sequence as coded in the genetic material), how such differences result in different three dimensional forms (tertiary structure), and how different 3-D structures lead to proteins function in various biological processes. An exact understanding of specific protein-function relation, therefore, requires the knowledge of 3-D structure of the corresponding protein. Unfortunately, we don't have 3-D structures of every protein. For example, there are only ~7,500 unique human proteins (compared to 21,000 possible proteins as mapped by the "Human Genome Project") deposited in the protein data bank (PDB), and keep in mind that many of these are small fragments not the whole proteins. The reason for such lacking is that isolating individual proteins is rather difficult and challenging processes experimentally. Therefore, scientists use computational tools such as homology modeling (or comparative modeling) to determine the 3-D structures of any given amino acid sequence using the data set of experimentally resolved protein structures in the PDB. In my PhD studies, I have developed on one such algorithm (D2) to describe proteins' structural propensity using the amino acid level details. Click HERE to see the details of this comparative modeling study.

Although static information of proteins' structures are useful, biological processes are naturally dynamic --- i.e., involving protein motion and structural changes. During my PhD, I have also worked on the dynamic aspect of protein functioning as to develop novel simulation protocols to investigate complex biological reactions in atomistic details. To that matter, I developed the adaptive steered molecular dynamics "ASMD" and multiple-branched ASMD (MB-ASMD), which explores the important nonequilibrium trajectories much more efficient than conventional MD and steered SMD. MB-ASMD can produce converged free energy profile of very complex bioprocesses by requiring 80% less trajectory data.

DNA compaction in the cell nucleus:

The 2-meter long DNA of eukaryotic cells is stored in the micron-sized nuclei (a compaction of ~50,000 folds) in a supercondensed form in which the double-stranded nucleic acid interacts with both core-histone proteins, linker histones, and monovalent (Na+/K+) and divalent (Mg2+) cations. Past research revealed that this enormous compaction is hierarchical with multiple levels of folding (see the "DNA in the CELL" image).

In the first level, approximately 147 base pairs of DNA is wrapped around an octamer of core-histone proteins about 1.7 turns to form what we call the "nucleosome". In the second level, with the addition of linker histones and various salts, multiple nucleosomes come together to form chromatin fibers. For a long time, scientist believed that a zigzag/solenoid structure that is 30 nanometers in thickness is the dominant form of chromatin at this level. However, recent research has shown that a less ordered and more dynamic form that is ~11 nm in thickness may also be abundant in the nucleus. My current research involves studying the the structure and dynamics of chromatin at the second level (green highlighted regions) to contribute this controversy.

I utilize biased and unconstrained Monte Carlo simulations to study structural and dynamical aspects of chromatin compaction under equilibrium and nonequilibrium conditions. Our chromatin model 1) takes the atomistic details of nucleosomes, 2) coarse-grains it to an irregular shape (that matches to the geometry of actual nucleosome), and 3) adds 300 point-charges (to account for electrostatic properties of the nucleosome) uniformly distributed on the surface of this irregular form. We then connect it with linker DNA (9 base pairs of DNA make a bead) to form nucleosome chains (green highlighted area on the DNA in the Cell picture above). Our strategy is kind of a bridge between very high resolution but expensive experiments such X-ray crystallography or Nuclear magnetic resonance (NMR) and low resolution experiments of the interphase chromosome (see the "DNA to Chromosome" image).

Page updated

Google Sites

Report abuse