Genetic Composition of Supercritical Branching Populations under Power Law Mutation Rates

Vianney Brouard
2024-02-13
Abstract:We aim at understanding the evolution of the genetic composition of cancer cell populations. To this aim, we consider a branching individual based model representing a cell population where cells divide, die and mutate along the edges of a finite directed graph $(V,E)$. The process starts with only one cell of trait $0$. Following typical parameter values in cancer cell populations we study the model under \textit{large population and power law mutation rates limit}, in the sense that the mutation probabilities are parameterized by negative powers of $n$ and the typical sizes of the population of interest are positive powers of $n$. Under \textit{non-increasing growth rate condition}, we describe the time evolution of the first-order asymptotics of the size of each subpopulation on the $log(n)$ time scale, as well as in the random time scale at which the initial population, resp. the total population, reaches the size $n^{t}$. In particular, such results allow for the perfect characterization of evolutionary pathways. Without imposing any conditions on the growth rates, we describe the time evolution of the order of magnitude of each subpopulation, whose asymptotic limits are positive non-decreasing piecewise linear continuous functions.
Probability,Populations and Evolution
What problem does this paper attempt to address?
### Problems the paper attempts to solve The paper aims to understand how the genetic composition of cancer cell populations evolves over time. Specifically, the author considers a branching individual - based model that represents a cell population in which cells divide, die, and mutate on the edges of a finite directed graph \( (V, E) \). The model starts with only one cell with the feature of 0 and studies the model behavior in the limit of large populations and power - law mutation rates. Specifically, the paper describes the first - order asymptotic behavior of the size of each sub - population in the evolution process on the \( \log(n) \) time scale and on the random time scale when the initial population or the total population reaches the size \( n^t \). These results help to perfectly characterize the evolutionary path. ### Key point summary 1. **Model background**: - The model assumes that cells have a phenotypic feature and the feature space \( V \) is finite. - The number of cells with each feature \( v \) at time \( t \) is represented as \( Z_v(t) \), and the whole process is represented as \( Z(t)=(Z_v(t))_{v\in V} \). - The initial condition is \( Z_v(0) = 1_{\{v = 0\}}\), that is, only cells with the feature of 0 exist. - A cell divides to produce two daughter cells, and the death rate and birth rate depend only on its phenotypic feature. - The mutation probability is encoded through the graph structure \( (V, E) \), and a mutation from feature \( v \) to feature \( u \) can occur only when \( (v, u)\in E \). 2. **Research objectives**: - To study the first - order asymptotic behavior of the size of each sub - population in the limit of large populations and power - law mutation rates. - To describe the evolution process of the size of each sub - population on the \( \log(n) \) time scale and the random time scale. - To characterize the evolutionary path, especially which mutation paths contribute to the growth of sub - populations. 3. **Main results**: - Under the condition of non - increasing growth rate (that is, \( \forall v\in V, \lambda(v)\leq\lambda(0) \)), the paper obtains the first - order asymptotic behavior of the size of each sub - population, whether on the random time scale or the deterministic time scale. - Without any growth rate assumptions, the paper studies the random exponential asymptotic behavior of the size of each sub - population. - The results show that the first - order asymptotic behavior of any sub - population size is completely determined by the randomness of the wild - type sub - population, rather than by the lineage dynamics of mutant cells or the randomness of the mutation process. - Further results show that only those paths with the minimum length and at most neutral mutations contribute to the growth of sub - populations. ### Biological interpretation 1. **Importance of mutation paths**: - The paper reveals which paths contribute to the growth of sub - populations by analyzing the length of different mutation paths and the number of neutral mutations. - Paths with the minimum length and at most neutral mutations contribute the most to the growth of sub - populations. 2. **Random and deterministic time scales**: - On the random time scale, the first - order asymptotic behavior of the sub - population size is completely determined by the randomness of the wild - type sub - population. - On the deterministic time scale, the asymptotic behavior of the sub - population size can be described by a deterministic mathematical formula. 3. **Impact of neutral mutations**: - The contribution of neutral mutations to sub - population growth is reflected by an additional \( \log(n) \) factor, which is captured in the asymptotic results. ### Conclusion Through mathematical models and asymptotic analysis, this paper deeply explores how the genetic composition of cancer cell populations evolves over time. The research results not only provide theoretical insights but also lay the foundation for developing new statistical methods to infer the evolutionary history of tumors.