Random Substitution-Insertion-Deletion (RSID) Model of Molecular Evolution with Alignment-free Parameter Estimation

TL;DR

This paper introduces a versatile, biologically accurate molecular evolution model called RSID, which enables alignment-free parameter estimation and reproduces observed indel length distributions in human pseudogenes.

Contribution

The paper presents the RSID model with heterogeneous rates and alignment-free parameter estimation, advancing the analysis of molecular evolution models.

Findings

Successfully reproduces indel length distribution in human pseudogenes

Provides a systematic, alignment-free parameter estimation method

Models heterogeneous rates and neighboring dependencies in evolution

Abstract

We present a comprehensive new framework for handling biologically accurate models of molecular evolution. This model provides a systematic framework for studying models of molecular evolution that implement heterogeneous rates, conservation of reading frame, differing rates of insertion and deletion, customizable parametrization of the probabilities and types of substitutions, insertions, and deletions, as well as neighboring dependencies. We have stated the model in terms of an infinite state Markov chain in order to maximize the number of applicable theorems useful in the analysis of the model. We use such theorems to develop an alignment-free parameter estimation technique. This alignment-free technique circumvents many of the nuanced issues related to alignment-dependent estimation. We then apply an implementation of our model to reproduce (in a completely alignment-free fashion) some observed results of Zhang and Gerstein (2003) regarding indel length distribution in human ribosomal protein pseudogenes.