I am working to further understanding of the applicability of Förster Resonance Energy Transfer (FRET) as a tool for probing biomolecular systems. With this tool, I examine the geometry and flexibility of nucleic acid polymers such as DNA and RNA in Lori Goldner’s single-molecule biophysics group.
Droplets for single-molecule confinement
One of the most common realizations of FRET experiment is by studying fluorescently-labelled freely-diffusing molecules under a common light microscope. While this approach can give details of kinetics and geometry for an ensemble, the inability to follow individual specimens precludes the study of dynamics of small numbers of molecules especially over long durations.
Building on a technique developed by Goldner at NIST, we use colloidal droplets as a means of containing fluorescently labelled molecules. By way of a novel particle tracking system, we can observe single-molecules contained within a droplet for minutes at a time, greatly extending the timescales accessible to single-molecule FRET.
Probabilistic analysis for better inference
By applying tools developed in the fields of statistics and machine learning, I seek to draw more meaningful and efficient inferences from experimental data. I use probabilistic graphical models along with Bayesian inference to construct generative models for photon data. This generative approach enables me to account for a variety of phenomena in a principled yet computationally tractable manner.
The time scales of events that can be studied with FRET is limited. From physical models we expect our observations to be perturbed by fast dynamics. In order to account for these effects we employ atomistic molecular dynamics simulations to shed light on these fast processes, allowing us to understand under which constraints our inference is valid.
Science should be open
Open access to hardware and software is essential for an open and reproducible scientific dialogue. In pursuit of this goal, I make the components of my experimental apparatus and analysis pipeline available for reuse and modification. One example of this is the freely available
fpga-timetagger, designed around an inexpensive off-the-shelf FPGA for use in single-photon counting applications. The design and associated software is available under an open-source license and serves as the primary data acquisition platform for my experimental work.
Additionally, all of my analysis code can be found on the group git repository If you are interested in any of the projects you find, don’t hesitate to send me an email.
- “Indocyanine dyes approach free rotation at 3’ terminus of A-RNA: A comparison with the 5’ terminus and consequences for FRET” (Journal of Physical Chemistry B)
- “MD Simulations and FRET Studies of Dye-Labeled RNA” (poster at Biophysical Society meeting, 2011)
- “From force-fields to photons: MD simulations of dye-labaled nucleic acids and Monte Carlo modelling of FRET” (poster at APS March Meeting, 2011)
- Gamari, et al. “Inexpensive hardware and software for photon statistics and correlation spectroscopy” arXiv:1206.3332 (submitted to American Journal of Physics)