Allele trees for the mother-dependent neutral mutations model and their scaling limits in the rare mutations regime
Airam Blancas, María Clara Fittipaldi, Saraí Hernández-Torres · 2025
Read Allele trees for the mother-dependent neutral mutations model and their scaling limits in the rare mutations regime by Airam Blancas, María Clara Fittipaldi, Saraí Hernández-Torres (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-20030.txt.
Introduction
Airam Blancas, María Fittipaldi, and Saraí Hernández-Torres study in 2025 allele family trees under a model where a mutant child never shares its mother's type. The model tracks a population split into a set number of types across generations. Past work covered the case of endless types, so the finite case extends that frame.
The team tracks family sizes for each type plus mutant counts as a chain over time. With a large start and rare mutation, the scaled chain settles to a new process form. They then lift that result to the family tree itself. The limit tree is built from ranked jumps of a random measure with a set intensity.
Catalog record
| CREATOR | Airam Blancas, María Clara Fittipaldi, Saraí Hernández-Torres |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 91 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-20030
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Allele trees for the mother-dependent neutral mutations model and their scaling limits in the rare mutations regime Airam Blancas, María Clara Fittipaldi, Saraí Hernández-Torres Abstract The mother-dependent neutral mutations model describes the evolution of a population across discrete generations, where neutral mutations occur among a finite set of possible alleles. In this model, each mutant child acquires a type different from that of its mother, chosen uniformly at random. In this work, we define a multitype allele tree associated with this model and analyze its scaling limit through a Markov chain that tracks the sizes of allelic subfamilies and their mutant descendants. We show that this Markov chain converges to a continuous-state Markov process, whose transition probabilities depend on the sizes of the initial allelic populations and those of their mutant offspring in the first allelic generation. As a result, the allele tree converges to a multidimensional limiting object, which can be described in terms of the universal allele tree introduced by Bertoin in [ 2 ] . 1 Introduction The Bienaymé-Galton-Watson (BGW) branching process is a classical model for the evolution of population size, in which each individual produces a random number of offspring independently of others, according to a fixed offspring distribution. The BGW process tracks the number of individuals across generations, indexed by discrete time, where generations represent the number of ancestral steps between an individual and the original ancestor. BGW processes have been extensively studied, and their variations extend the model’s applicability beyond population size, allowing for the modeling of a wide range of phenomena that shape population structure. Among the key forces driving population change, gene mutations play a crucial role. Given a specific gene, mutations give rise to different versions, referred to as alleles . Research in computational biology and genetics has highlighted the need for models that consider a finite number of alleles; see [ 9 , 11 , 14 ] . The mother-dependent neutral mutations model , introduced in [ 3 ] , addresses this need by modeling the genealogy of a population undergoing neutral mutations with a finite number of alleles (see Subsection 1.1 ). In this work, we analyze its long-term allelic structure in the following sense. Alleles classify the population’s genealogy according to genetic type, leading to a partition into allelic subfamilies , which are encoded in a multitype allele tree (defined in Section 4 ). Our main focus is on the allele tree and its scaling limit. Our analysis considers a population modeled by mother-dependent neutral mutations, where population growth is in the critical regime, the mutation rate is small, and the number of original ancestors is large (see Hypotheses (H 1 ), (H 2 ), and (H 3 ) below). Before stating our main theorem, we first define the mother-dependent neutral mutations model as a multitype BGW process. 1.1 Mother-dependent neutral mutations model The mother-dependent neutral mutations model describes the evolution of a population with a finite number of alleles, where random mutations alter the genotype of individuals without affecting their offspring distribution–and hence without influencing population size dynamics. Because of this property, these mutations are referred to as neutral . Children who inherit the same allele as their mother will be referred to as clones , while those who acquire a different allele will be called mutants . In the mother-dependent neutral mutations model, population size follows a standard BGW process, and mutations arise along the ancestral lines of the population. A parameter r\in(0,1) controls the mutation probability , with mutations occurring randomly within the population and independently of one another. Whenever a mutation event occurs, the mutant child does not inherit the mother’s allele but instead acquires a different allele chosen uniformly at random from a finite set of possible alleles. Throughout this work, let d be the number of alleles, which we label as [d]\coloneqq\{1,\ldots,d\} . Then, the joint distribution of mutant children follows a multinomial distribution, which we introduce next. We denote by \boldsymbol{\mu}=(\boldsymbol{\mu}_{i},i\in[d]) the offspring distribution of our population, where each \boldsymbol{\mu}_{i} is a measure valued on \mathbb{N}_{0}^{d} . The measure \boldsymbol{\mu}_{i} specifies the offspring distribution (by type) of an individual of type i . Since we assume that the mutations in our model are neutral, any individual, regardless of type, follows a common offspring distribution \mu for its total number of children . We thus assume that \mu is a probability measure supported on \mathbb{N}_{0} , satisfying \mu(0)+\mu(1)<1 (to avoid trivial cases). For i\in[d] , the measure \boldsymbol{\mu}_{i} defines the distribution of the random vector \boldsymbol{\xi}^{i} , corresponding to the offspring of an individual of type i . According to our description above, the probability that an individual of type i begets \mathbf{v}=(\mathbf{v}(1),\ldots,\mathbf{v}(d))\in\mathbb{N}_{0}^{d} children is \begin{split}\boldsymbol{\mu}_{i}(\mathbf{v})&=\mathbb{P}_{i}(\boldsymbol{\xi}^{i}=\mathbf{v})\\ &=\mathbb{P}_{\mathbf{e}_{i}}\left[(\boldsymbol{\xi}^{i}{(1)},\dots,\boldsymbol{\xi}^{i}{(d)})=(\mathbf{v}(1),\dots,\mathbf{v}(d))\right]\\ &=\mu(|\mathbf{v}|)\binom{|\mathbf{v}|}{\mathbf{v}(1),\dots,\mathbf{v}(d)}(1-r)^{\mathbf{v}(i)}\prod\limits_{j\neq i}\left(\dfrac{r}{d-1}\right)^{\mathbf{v}(j)}.\\ \end{split} (1) In the equation above, \mu_{i}({\mathbf{v}}) is the probability that an individual of type i gives birth to |\mathbf{v}| children, with \mathbf{v}(j) of them being of type j . Equation ( 1 ) implies that, among the \mu(|\mathbf{v}|) children, each undergoes a mutation with probability r\in(0,1) , independently of the others. The types of mutant children are assigned according to a multinomial distribution. Then the random variable \xi^{(+)}\overset{d}{=}\xi^{(+,i)}\coloneqq\sum_{j=1}^{d}\boldsymbol{\xi}^{i}(j),\qquad i\in[d], (2) follows distribution \mu , which is the same for all types of individuals. In the mother-dependent neutral mutations model, the population is described by a d -type BGW process \{\mathbf{Y}=(\mathbf{Y}_{k},k\in\mathbb{N}_{0});\mathbb{P}_{\mathbf{a}}^{\boldsymbol{\mu},r}\} . Here, we indicate that the process \mathbf{Y} has \boldsymbol{\mu} as its offspring distribution (as defined in ( 1 )), with mutation rate r\in[0,1] , and starts with \mathbf{a}\in\mathbb{N}_{0}^{d} individuals. That is, \mathbf{a}(i)\in\mathbb{N}_{0} represents the initial number of individuals of type i . Since the relevant parameter for the measure \boldsymbol{\mu} is the mutation rate r , we will use the notation \mathbb{P}_{\mathbf{a}}^{r} . For each k\in\mathbb{N}_{0} , \mathbf{Y}_{k} describes the allelic configuration of the population in the k -th generation (also referred to as the k -th level), so that for each i\in[d] , \mathbf{Y}_{k}(i) denotes the number of individuals of type i in the k -th level. The process \mathbf{Y}\coloneqq(\mathbf{Y}_{k},k\in\mathbb{N}_{0}) is a discrete-time Markov chain, recursively defined for each j\in[d] by \mathbf{Y}_{0}(j)=\mathbf{a}(j)\qquad\text{and}\qquad\mathbf{Y}_{k+1}(j)=\sum_{i=1}^{d}\sum_{\ell=1}^{\mathbf{Y}_{k}(i)}\boldsymbol{\xi}_{k+1,\ell}^{i}{(j)},\qquad\text{ for all }k\in\mathbb{N}_{0}. (3) In the definition above, \boldsymbol{\xi}^{i}_{k+1,\ell}(j) denotes the number of children of type j in the (k+1) -th level begotten by the \ell -th individual of type i in level k . The random vectors (\boldsymbol{\xi}^{i}_{k,\ell})_{k\in\mathbb{N}_{0},\ell\in\mathbb{N}} are independent copies of the random vector \boldsymbol{\xi}^{i} defined in ( 1 ). X X X X X X X X X 3 3 1 1 2 2 3 2 1 1 Figure 1 : Left: A realization of the mother-dependent neutral mutation model with d=3 types. The types 1 , 2 and 3 are represented in green, orange, and purple, respectively. Mutations are indicated by marks on the edges, and the representative of each allelic subfamily is shown in a bolder shade. Right: Multitype allele tree associated with the colored genealogical tree on the left. 1.2 Main result on multiallele trees A multitype allele tree is a process \mathscr{A}=((\mathcal{A}_{u},\mathcal{C}_{u},\boldsymbol{d}_{u}):u\in\mathbb{U}) (4) indexed by the Ulam-Harris tree \mathbb{U} (introduced in Subsection 2.2 ). Each u\in\mathbb{U} represents an allelic subfamily , where \mathcal{A}_{u} denotes the size of the subfamily, \mathcal{C}_{u} indicates the allele type, and \boldsymbol{d}_{u} is a vector containing