University of British Columbia
Mathematical modeling and analysis of biological shape heterogeneity and cellular growth
Abstract
dc:descriptionRecent advances in microscopy and imaging techniques have generated an unprecedented volume of biological data, from the molecular to the organismal level. New microscopy techniques allow researchers to capture biological shape heterogeneity at high resolution, and shape dynamics at high frequency. This increase in biological data raises challenges for comparing shapes and modeling their dynamics. This dissertation develops theoretical and computational methods for analyzing biological shapes, and mathematical models for describing shape heterogeneity, across four projects. The first two projects bridge shape variations with growth mechanisms: first, we uncover how molecular mechanisms controlling ribosome populations can produce experimentally observed oscillations in mouse liver size. More precisely, we introduce a dynamical system including the creation and recycling of ribosomes, and show that optimal strategies to maximize and maintain protein production can generate such oscillations. Second, we analyze atomic force microscopy time series of Mycobacterium smegmatis to characterize interactions between the bacterium's growth, morphological pattern dynamics, and division site selection. More precisely, we statistically confirm and study its stable wave-form morphological patterns, and show that its biphasic polar growth influences where it divides. To model the cell shape dynamics, we introduce a mathematical model reproducing the emergence and dynamics of these surface features. In the third project, we model the emergence of geometrical defects in the fruit fly meiotic spindle, the microtubule-based apparatus responsible for chromosome segregation. We model the positions of microtubules as a spatial birth-death process and represent spindle defects as large gaps between consecutive microtubules. We show that the time of defect onset is exponentially distributed in a specific asymptotic limit, and compute its expected value explicitly. The last project introduces a class of new metrics on shapes, derived from optimal transport theory, that can capture certain geometric properties and symmetries. We study their theoretical properties and robustness, while uncovering their underlying low-dimensional structure. We leverage this structure to develop an efficient algorithm for computing the metrics. In particular, we introduce the Chiral Gromov-Wasserstein distance, adapted to compare chiral molecules and proteins, and validate our framework with numerical experiments on chiral objects.
Degree
thesis:*- Name thesis:degree_name
- Doctor of Philosophy - PhD
- Level thesis:degree_level
- doctoral
- Discipline thesis:degree_discipline
- Mathematics
- Grantor dc:publisher
- University of British Columbia
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Soubrier, Clément
- Contributors dc:contributor
-
- Dao Duc, Khanh
Rights
dc:rights- Statement dc:rights
-
- Attribution 4.0 International
- Language dc:language
- eng
Identifiers
dc:identifier.*- Handle dc:identifier
- http://hdl.handle.net/2429/95438