An F-actin point pattern cluster analysis with triangulation algorithms

Download files
Access & Terms of Use
open access
Copyright: Tran, Jason
Altmetric
Abstract
The lamellipodia is a dense, branched actin network that dictates cellular processes such as cell adhesion and migration. Although the actin cytoskeleton has been the focus of many SMLM methods, no attempt has been made to image and directly quantify the F-actin distribution in the lamellipodia alone. In this thesis, I aimed to design a novel point pattern analysis that is based on Delaunay triangulation and quantify the F-actin distribution and organisation in the lamellipodia. In Chapter 3, Ripley’s K-function is applied to SMLM point pattern data of Cav-1. Here I demonstrate that Ripley’s K-function is only suitable for punctate clusters and that one of the main limitations is that it describes the global degree of clustering and it is limited to detecting circular clusters. I implement Ripley’s K-function to Cav-1, which is known to cluster to form caveolin pockets in the membrane and show that Cav-1 clustering distribution was not affected by sterol enrichment, which was on par with the TEM data. In Chapter 4, a Delaunay triangulation-based analysis is designed to overcome the limitations of current SMLM analytical methodologies and suitable for quantifying dense lamellipodia F-actin SMLM data. The analysis includes segmenting the cell edge, distinguishing that edge from those generated artificially by the image boundary (which is later used to identify the lamellipodia in chapter 5), and further segmenting and quantifying interior clusters. While subjecting simulated clusters to various background point/noise point densities, the mean number of simulated clusters and the size and density were in close approximation of the pre-determined simulated values. This demonstrated that the analysis is highly robust. In Chapter 5, the Delaunay triangulation-based analysis is implemented for analysis of SMLM data of F-actin in the lamellipodia during its intermediary retraction and extension phases. The global and local density distributions as well as the organisation of the lamellipodia are investigated. I show that during the retraction phase the lamellipodia is initially a uniformly dense network and transitions into a highly dense and clustered structure. This transition in F-actin network organisation is reversed during the protrusion phase following from a full retraction. Altogether (Chapter 6), this work demonstrates that the Delaunay triangulation-based analysis method is suitable to characterise, segment, and quantify highly dense point pattern data from SMLM. The work also provides a novel insight into the subtle F-actin clustering and spatial arrangement within the lamellipodia during the retraction and extension phases, which was previously inaccessible by conventional diffraction-limited microscopy studies.
Persistent link to this record
Link to Publisher Version
Link to Open Access Version
Additional Link
Author(s)
Tran, Jason
Supervisor(s)
Gaus, Katharina
Nieves, Daniel
Creator(s)
Editor(s)
Translator(s)
Curator(s)
Designer(s)
Arranger(s)
Composer(s)
Recordist(s)
Conference Proceedings Editor(s)
Other Contributor(s)
Corporate/Industry Contributor(s)
Publication Year
2018
Resource Type
Thesis
Degree Type
PhD Doctorate
UNSW Faculty
Files
download public version.pdf 8.02 MB Adobe Portable Document Format
Related dataset(s)