Multiplexed Super-Resolution Imaging : Design of deoxyribonucleic acid (DNA) oligonucleotides for multiplexed imaging on DNA origami nanostructures
Haider, Belquis (2021)
2021
Master's Programme in Photonics Technologies
Tekniikan ja luonnontieteiden tiedekunta - Faculty of Engineering and Natural Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2021-11-24
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202111238615
https://urn.fi/URN:NBN:fi:tuni-202111238615
Tiivistelmä
This project aims to investigate the possibility of multiplexing and achieving a 20-nanometer resolution in an existing setup using single-molecule localisation microscopy methods. In addition, this thesis aims to establish a protocol to design and synthesise optimised DNA origami nanostructures in-house. These nanostructures are of interest since they have been extensively used as nanoscale rulers in previous super-resolution applications.
The technique covered is DNA-PAINT (DNA point-accumulation in nanoscale topology). First, DNA nanostructures are designed in silico and simulated to ensure their mechanical stability. Various considerations pertaining to the optimisation of nanostructure synthesis are discussed and adopted to improve yield and suitability in DNA-PAINT microscopy experiments. The thesis then describes the different parameters affecting the quality of images and achieved spatial resolutions, with the signal to noise ratio and the number of acquired photons being the most significant. In addition, different sources of error such as drift are discussed along with minimisation techniques. This report also details the simulation process and results under different laser power densities and imager concentrations, illustrating that a 20-nanometre spatial resolution is achievable under conditions where the imager concentration and power density are increased at the cost of temporal resolution. As for the experimental results, they depict the possibility of multiplexing and distinguishing structures based on engineering the binding sites on DNA nanostructures. Although not visible at first, the nanostructures can be distinguished with image post-processing and filtering techniques. A MATLAB code has been developed as part of this project to automatically filter and facilitate analysis for picked structures. Data analysis of rendered and filtered structures shows that the structures with a total of 20 and 40 binding sites exhibit 19.6 and 44 binding events respectively. This is within a 2% and 10% margin of error. However, the results from the structure with 30 binding sites in total exhibited merely an average of about 10 binding events per structure. This value is in agreement with the expected value for the central sites indicating either incorporation issues of the sites in the periphery of the structure or low sensitivity in detecting the lower intensity spots. Analysis of the maximum binding events per structure achieved from a given experiment showed an average of 25 binding events which is still lower than the anticipated value. Nonetheless, the analysed structures still appeared to exhibit expected properties in terms of brightness and spatial resolution. Other parameters that were investigated in this project include the bright, dark, and decay times along with their dependence on imager concentration, availability of sites (binding frequency), and laser power.
The technique covered is DNA-PAINT (DNA point-accumulation in nanoscale topology). First, DNA nanostructures are designed in silico and simulated to ensure their mechanical stability. Various considerations pertaining to the optimisation of nanostructure synthesis are discussed and adopted to improve yield and suitability in DNA-PAINT microscopy experiments. The thesis then describes the different parameters affecting the quality of images and achieved spatial resolutions, with the signal to noise ratio and the number of acquired photons being the most significant. In addition, different sources of error such as drift are discussed along with minimisation techniques. This report also details the simulation process and results under different laser power densities and imager concentrations, illustrating that a 20-nanometre spatial resolution is achievable under conditions where the imager concentration and power density are increased at the cost of temporal resolution. As for the experimental results, they depict the possibility of multiplexing and distinguishing structures based on engineering the binding sites on DNA nanostructures. Although not visible at first, the nanostructures can be distinguished with image post-processing and filtering techniques. A MATLAB code has been developed as part of this project to automatically filter and facilitate analysis for picked structures. Data analysis of rendered and filtered structures shows that the structures with a total of 20 and 40 binding sites exhibit 19.6 and 44 binding events respectively. This is within a 2% and 10% margin of error. However, the results from the structure with 30 binding sites in total exhibited merely an average of about 10 binding events per structure. This value is in agreement with the expected value for the central sites indicating either incorporation issues of the sites in the periphery of the structure or low sensitivity in detecting the lower intensity spots. Analysis of the maximum binding events per structure achieved from a given experiment showed an average of 25 binding events which is still lower than the anticipated value. Nonetheless, the analysed structures still appeared to exhibit expected properties in terms of brightness and spatial resolution. Other parameters that were investigated in this project include the bright, dark, and decay times along with their dependence on imager concentration, availability of sites (binding frequency), and laser power.