Quantifying Transcriptional Dynamics and Their Effects on Genetic Motifs from Live Cell Fluorescence Microscopy
Häkkinen, Antti (2016)
Häkkinen, Antti
Tampere University of Technology
2016
Teknis-taloudellinen tiedekunta - Faculty of Business and Technology Management
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:ISBN:978-952-15-3702-8
https://urn.fi/URN:ISBN:978-952-15-3702-8
Tiivistelmä
Advances in measurement techniques based on fluorescent tagging have enabled visualizing individual transcripts and proteins over time as they are produced in live cells. Such methods are critical in understanding how genes and genetic networks function, how they respond to external signals, such as stress conditions and temperature changes, and how cellular aging and diseases can affect their performance. This is relevant, as the functioning of genes and genetic networks affects the survival of cells and cell populations.
However, as cellular processes are complex and inherently stochastic, statistical signal processing methods are required to pre-process, analyze, and interpret the results from the measurement data. This stems from various traits of the data: averages poorly describe the behavior of multimodal populations; confidence estimates and hypothesis testing must be used to compare results, as they feature significant variability; and large data sets are required such that comparison can be made with sufficient confidence, rendering manual quantification excessively laborious. In addition, when combined with stochastic models, such methods can be used to extract information, which cannot be directly measured with current techniques. Meanwhile, the methods must be carefully designed, in order to avoid hidden assumptions which obstruct the objective quantification and comparison of the results. Finally, the methods should be made robust against errors characteristic to the measurement system or propagating from the earlier stages of the analysis.
Here, methods were developed in order to enhance the amount and the quality of the information which can be extracted from single-molecule measurements of live cells. In particular, methods for estimating RNA numbers and RNA production intervals from static images of cell populations and from time series of images of growing cells were first established. Next, methods for estimating the subprocesses of transcription were developed, as these processes cannot be directly measured in live cells. Computer simulations and live single-RNA measurements were used to demonstrate the reliability and performance of the new methods, indicating that the methods can adapt to different measurement settings and can be applied to other similar dynamical estimation problems. Finally, computer simulations of genetic networks are used to demonstrate that the accuracy of such methods is paramount, as, in the dynamical ranges extracted from measurement, changes in gene expression dynamics have implications on the behavior of genetic networks, which are reflected on the behavior of individual cells and of cell populations as a whole.
The outcomes of this thesis respond to the demand of carefully designed statistical methods for accurate and unbiased quantification and comparison of cellular processes. Advances in such methods are necessary in order to generate new insight on the dynamics and the regulatory mechanisms of gene expression from single-molecule, single-cell measurements in live cells. The methods and the findings presented here will be critical for the success of such studies, contributing toward understanding how changes in gene expression patterns influence the cellular aging, stress, and diseases.
However, as cellular processes are complex and inherently stochastic, statistical signal processing methods are required to pre-process, analyze, and interpret the results from the measurement data. This stems from various traits of the data: averages poorly describe the behavior of multimodal populations; confidence estimates and hypothesis testing must be used to compare results, as they feature significant variability; and large data sets are required such that comparison can be made with sufficient confidence, rendering manual quantification excessively laborious. In addition, when combined with stochastic models, such methods can be used to extract information, which cannot be directly measured with current techniques. Meanwhile, the methods must be carefully designed, in order to avoid hidden assumptions which obstruct the objective quantification and comparison of the results. Finally, the methods should be made robust against errors characteristic to the measurement system or propagating from the earlier stages of the analysis.
Here, methods were developed in order to enhance the amount and the quality of the information which can be extracted from single-molecule measurements of live cells. In particular, methods for estimating RNA numbers and RNA production intervals from static images of cell populations and from time series of images of growing cells were first established. Next, methods for estimating the subprocesses of transcription were developed, as these processes cannot be directly measured in live cells. Computer simulations and live single-RNA measurements were used to demonstrate the reliability and performance of the new methods, indicating that the methods can adapt to different measurement settings and can be applied to other similar dynamical estimation problems. Finally, computer simulations of genetic networks are used to demonstrate that the accuracy of such methods is paramount, as, in the dynamical ranges extracted from measurement, changes in gene expression dynamics have implications on the behavior of genetic networks, which are reflected on the behavior of individual cells and of cell populations as a whole.
The outcomes of this thesis respond to the demand of carefully designed statistical methods for accurate and unbiased quantification and comparison of cellular processes. Advances in such methods are necessary in order to generate new insight on the dynamics and the regulatory mechanisms of gene expression from single-molecule, single-cell measurements in live cells. The methods and the findings presented here will be critical for the success of such studies, contributing toward understanding how changes in gene expression patterns influence the cellular aging, stress, and diseases.
Kokoelmat
- Väitöskirjat [4891]