A Method for Testing Random Spatial Models on Nuclear Object Distributions

Abstract

The cell nucleus is a structurally complex and dynamic organelle ensuring key biological functions. Complex relationships between nuclear structure and functions require a better understanding of the three-dimensional organization of the genome and of the subnuclear compartments. Quantitative image analysis coupled with spatial statistics and modeling is a relevant approach to address these questions. In this chapter, we describe a step-by-step procedure to process images and to test a spatial random model for the distribution of nuclear objects using chromocenters as an example. More elaborate models can be designed on the basis of the random model by introducing additional and more complex constraints to better fit observations and to question determinants of these spatial organizations.

2D morphometric analysis of Arabidopsis thaliana nuclei reveals characteristic profiles of different cell types and accessions

“…The RHF and HX parameters have been used since then to measure the dynamics of heterochromatin compaction during development and in response to biotic and abiotic stress conditions (Tessadori et al 2007a, b;2009;van Zanten et al 2010van Zanten et al , 2011Pecinka et al 2010;Bourbousse et al 2015). A more sophisticated method was developed using 3D image processing technology for measuring the relative heterochromatin volume (RHV) and the position and distances between chromocenters (Andrey et al 2010;Poulet et al 2015Poulet et al , 2017Ashenafi and Baroux, 2018;Desset et al 2018;Arpòn et al 2018). However, this accurate computational analysis of the 3D nuclear phenotype is in some cases less applicable for morphometric analysis, due to the time-consuming acquisition and analysis of the confocal images.…”

Heterogeneity and its multiscale integration in plant morphogenesis

“…However, dealing with heterogeneity also introduces new image analysis problems, in particular when it comes to identify principles of organization from noisy spatial image data. Methods based on image normalization and spatial statistics are emerging to address such problems [65,66]. Similarly, it can be anticipated that stochastic modeling approaches will be promoted in the coming years, as deterministic models have shown their limits when addressing noise and heterogeneity in various processes such as cell division [67].…”

Image analysis workflows to reveal the spatial organization of cell nuclei and chromosomes

“…The field of chromatin, chromosome and nuclear organization studies would greatly benefit from the broader deployment of image processing-based quantitative analyses [2,3]. Several tools and packages have been developed in the past years based on open-source software, including for the 3D spatial analysis of nuclear organization [7][8][9][10][11]. Yet, a major hurdle for most 'biology-only' oriented labs is the lack of computational expertise for customizing the image processing scripts, for large data handling, the lack of template workflows, or a combination thereof.…”