dorsal/arxiv
View SchemaLimitations of Markov chain Monte Carlo algorithms for Bayesian Inference of phylogeny
| Authors | Elchanan Mossel, Eric Vigoda |
|---|---|
| Categories | |
| ArXiv ID | q-bio/0505002 |
| URL | https://arxiv.org/abs/q-bio/0505002 |
| DOI | 10.1214/105051600000000538 |
| Journal | Annals of Applied Probability 2006, Vol. 16, No. 4, 2215-2234 |
Abstract
Markov chain Monte Carlo algorithms play a key role in the Bayesian approach to phylogenetic inference. In this paper, we present the first theoretical work analyzing the rate of convergence of several Markov chains widely used in phylogenetic inference. We analyze simple, realistic examples where these Markov chains fail to converge quickly. In particular, the data studied are generated from a pair of trees, under a standard evolutionary model. We prove that many of the popular Markov chains take exponentially long to reach their stationary distribution. Our construction is pertinent since it is well known that phylogenetic trees for genes may differ within a single organism. Our results shed a cautionary light on phylogenetic analysis using Bayesian inference and highlight future directions for potential theoretical work.
{
"annotation_id": "6406d15c-72f5-470e-8e7e-d1d749bb17c7",
"date_created": "2026-03-02T18:01:32.243000Z",
"date_modified": "2026-03-02T18:01:32.243000Z",
"file_hash": "46d55fe1dc6cd5437bd81f29033bd6c136f9d0b0885178191af69c0be6abbbdd",
"private": false,
"record": {
"abstract": "Markov chain Monte Carlo algorithms play a key role in the Bayesian approach\nto phylogenetic inference. In this paper, we present the first theoretical work\nanalyzing the rate of convergence of several Markov chains widely used in\nphylogenetic inference. We analyze simple, realistic examples where these\nMarkov chains fail to converge quickly. In particular, the data studied are\ngenerated from a pair of trees, under a standard evolutionary model. We prove\nthat many of the popular Markov chains take exponentially long to reach their\nstationary distribution. Our construction is pertinent since it is well known\nthat phylogenetic trees for genes may differ within a single organism. Our\nresults shed a cautionary light on phylogenetic analysis using Bayesian\ninference and highlight future directions for potential theoretical work.",
"arxiv_id": "q-bio/0505002",
"authors": [
"Elchanan Mossel",
"Eric Vigoda"
],
"categories": [
"q-bio.PE",
"q-bio.GN"
],
"doi": "10.1214/105051600000000538",
"journal_ref": "Annals of Applied Probability 2006, Vol. 16, No. 4, 2215-2234",
"title": "Limitations of Markov chain Monte Carlo algorithms for Bayesian Inference of phylogeny",
"url": "https://arxiv.org/abs/q-bio/0505002"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "d4be443c-3107-42f2-801c-21cf5a3ace6a",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}