HAL: hal-00328181, version 1

DOI: 10.1016/j.csda.2008.07.007

Detailed view

Export this paper BibTeX - EndNote - TEI - RefWorks -

Computational Statistics & Data Analysis (2009) 9999

Bayesian hidden Markov Model for DNA segmentation : A prior sensitivity analysis

Darfiana Nur (  Corresponding author ) 1, David Allingham 2, Judith Rousseau 3, 4, Kerrie Mengersen 5, Ross Mcvinish 5

For the Darfiana Nur,David Allingham, Judith Rousseau , Kerrie L. Mengersen, Ross McVinishd collaboration(s)

(2009)

The focus of this paper is on the sensitivity to the specification of the prior in a hidden Markov model describing homogeneous segments of DNA sequences. An intron from the chimpanzee α-fetoprotein gene, which plays an im- portant role in embryonic development in mammals is analysed. Three main aims are considered : (i) to assess the sensitivity to prior specification in Bayesian hidden Markov models for DNA sequence segmentation; (ii) to examine the impact of replacing the standard Dirichlet prior with a mixture Dirichlet prior; and (iii) to propose and illus- trate a more comprehensive approach to sensitivity analysis, using importance sampling. It is obtained that (i) the posterior estimates obtained under a Bayesian hidden Markov model are indeed sensitive to the specification of the prior distributions; (ii) compared with the standard Dirichlet prior, the mixture Dirichlet prior is more flexible, less sensitive to the choice of hyperparameters and less constraining in the analysis, thus improving posterior estimates; and (iii) importance sampling was computationally feasible, fast and effective in allowing a richer sensitivity analysis.

1:	School of Mathematical and physical Sciences
	University of Newcastle
2:	ARC center of excellence for Complex Dynamic Systems and Control
	ARC center of excellence for complex dunamic systems and control
3:	CEntre de REcherches en MAthématiques de la DEcision (CEREMADE)
	CNRS : UMR7534 – Université Paris IX - Paris Dauphine
4:	Centre de Recherche en Économie et Statistique (CREST)
	INSEE – École Nationale de la Statistique et de l'Administration Économique
5:	school of mathematical sciences
	Queensland University of Technology

Subject

:

Mathematics/Statistics Statistics/Statistics Theory

Keyword(s): DNA sequence – hidden Markov model – Bayesian model – sensitivity analysis – α-fetoprotein – Markov chain Monte Carlo – importance sampling.

hal-00328181, version 1

http://hal.archives-ouvertes.fr/hal-00328181

oai:hal.archives-ouvertes.fr:hal-00328181

From: Judith Rousseau <>

Submitted on: Friday, 10 October 2008 22:10:25

Updated on: Saturday, 14 February 2009 17:00:07