Path integral formulation and Feynman rules for phylogenetic branching models

A dynamical picture of phylogenetic evolution is given in terms of Markov models on a state space, comprising joint probability distributions for character types of taxonomic classes. Phylogenetic branching is a process which augments the number of taxa under consideration, and hence the rank of the...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2005-10
Main Authors: Jarvis, P D, Bashford, J D, Sumner, J G
Format: Article
Language:English
Subjects:
Combinatorial analysis
Covariance
Diffusion rate
Evolution
Integrals
Leaves
Markov chains
Operators (mathematics)
Perturbation methods
Perturbation theory
Phylogenetics
Tensors
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A dynamical picture of phylogenetic evolution is given in terms of Markov models on a state space, comprising joint probability distributions for character types of taxonomic classes. Phylogenetic branching is a process which augments the number of taxa under consideration, and hence the rank of the underlying joint probability state tensor. We point out the combinatorial necessity for a second-quantised, or Fock space setting, incorporating discrete counting labels for taxa and character types, to allow for a description in the number basis. Rate operators describing both time evolution without branching, and also phylogenetic branching events, are identified. A detailed development of these ideas is given, using standard transcriptions from the microscopic formulation of nonequilibrium reaction-diffusion or birth-death processes. These give the relations between stochastic rate matrices, the matrix elements of the corresponding evolution operators representing them, and the integral kernels needed to implement these as path integrals. The `free' theory (without branching) is solved, and the correct trilinear `interaction' terms (representing branching events) are presented. The full model is developed in perturbation theory via the derivation of explicit Feynman rules which establish that the probabilities (pattern frequencies of leaf colourations) arising as matrix elements of the time evolution operator are identical with those computed via the standard analysis. Simple examples (phylogenetic trees with 2 or 3 leaves), are discussed in detail. Further implications for the work are briefly considered including the role of time reparametrisation covariance.
ISSN:2331-8422
DOI:10.48550/arxiv.0411047