Filtering by
- Member of: Theses and Dissertations
Single molecule FRET experiments are important for studying processes that happen on the molecular scale. By using pulsed illumination and collecting single photons, it is possible to use information gained from the fluorescence lifetime of the chromophores in the FRET pair to gain more accurate estimates of the underlying FRET rate which is used to determine information about the distance between the chromophores of the FRET pair. In this paper, we outline a method that utilizes Bayesian inference to learn parameter values for a model informed by the physics of a immobilized single-molecule FRET experiment. This method is unique in that it combines a rigorous look at the photophysics of the FRET pair and a nonparametric treatment of the molecular conformational statespace, allowing the method to learn not just relevant photophysical rates (such as relaxation rates and FRET rates), but also the number of molecular conformational states.
A statistical method is proposed to learn what the diffusion coefficient is at any point in space of a cell membrane. The method used bayesian non-parametrics to learn this value. Learning the diffusion coefficient might be useful for understanding more about cellular dynamics.
Bdellovibrio bacteriovorus (B. bacteriovorus) is a predatory bacterium that preys on other gram-negative bacteria. In order to survive and reproduce, B. bacteriovorus invades the periplasm of other bacterial cells creating the potential for it to act as a “living antibiotic”. In this work, a comparison was made between the rates of predation of B. bacteriovorus in vitro and in vivo. In vitro, the behavior of B. bacteriovorus was examined in the presence of prey. In vivo, the behavior of B. bacteriovorus was examined in the presence of prey and a living host, Caenorhabditis elegans (C. elegans). C. elegans were infected with Escherichia coli (E. coli) and treated with B. bacteriovorus. In previous studies that analyzed B. bacteriovorus in vitro, a decrease in concentrations of bacteria has been observed after introduction of B. bacteriovorus. In vivo, B. bacteriovorus were found to not have a net reduction of E. coli but to reproducibly raise the level of fluctuations in E. coli concentrations.
Bdellovibrio bacteriovorus (BB) is a gram negative predatory bacteria that uses other gram negative bacteria to proliferate non-binarily. Due to the predatory nature of BB researchers have proposed to use it as a potential biocontrol agent against other gram negative bacteria. The in vivo effect of predatory bacteria on a living host lacks thorough investigation. This paper explores BB inside and outside of the C. elegans. BB acts internally by pre- infecting C. elegans with E. coli and then treating the worms with BB. After BB treatment worm survivavbility increased and morbidity decreased. Ex- ternally, BB modulated the environment around the nematode which reduced infection rates and increased nematode lifespan and survivability. Together, the internal and external results suggest BB has the capability to act as a living antibiotic acting topically and internally to reduce infection rates.
Electron Multiplying Charge Coupled Device (EMCCD) cameras are widely used for fluorescence microscopy experiments. However, the quantitative determination of biological parameters uniquely depends on characteristics of the unavoidably inhomogenous illumination profile as it gives rise to an image. It is therefore of interest to learn this inhomogenous illumination profiles that can dramatically vary across images alongside the camera parameters though a detailed camera model. In this manuscript we create a detailed model to learn inhomogeneous illumination profiles as well as learn all associated camera parameters. We achieve this within a Bayesian paradigm allowing us to determine full distributions over the unknowns.
This thesis proposes a graph based neural network architecture, SwarmNet, for learning the swarming behaviors of multi-agent systems. Given observation of only the trajectories of an expert multi-agent system, the SwarmNet is able to learn sensible representations of the internal low-level interactions on top of being able to approximate the high-level behaviors and make long-term prediction of the motion of the system. Challenges in scaling the SwarmNet and graph neural networks in general are discussed in detail, along with measures to alleviate the scaling issue in generalization is proposed. Using the trained network as a control policy, it is shown that the combination of imitation learning and reinforcement learning improves the policy more efficiently. To some extent, it is shown that the low-level interactions are successfully identified and separated and that the separated functionality enables fine controlled custom training.
Several experimental measurements can probe diffusion coefficients at the single-molecule and bulk level. The target of this thesis is on single-molecule methods, which can assess diffusion coefficients at the individual molecular level. For instance, super resolution methods like stochastic optical reconstruction microscopy (STORM) and photo activated localization microscopy (PALM), have a high spatial resolution with the cost of lower temporal resolution. Also, there is a different group of methods, such as MINFLUX, multi-detector tracking, which can track a single molecule with high spatio-temporal resolution. The problem with these methods is that they are only applicable to very diluted samples since they need to ensure existence of a single molecule in the region of interest (ROI).
In this thesis, the goal is to have the best of both worlds by achieving high spatio-temporal resolutions without being limited to a few molecules. To do so, one needs to refocus on fluorescence correlation spectroscopy (FCS) as a method that applies to both in vivo and in vitro systems with a high temporal resolution and relies on multiple molecules traversing a confocal volume for an extended period of time. The difficulty here is that the interpretation of the signal leads to different estimates for the kinetic parameters such as diffusion coefficients based on a different number of molecules we consider in the model. It is for this reason that the focus of this thesis is now on using Bayesian nonparametrics (BNPs) as a way to solve this model selection problem and extract kinetic parameters such as diffusion coefficients at the single-molecule level from a few photons, and thus with the highest temporal resolution as possible.