File Download

There are no files associated with this item.

Conference Paper: Bayesian Analysis of EST Data with Multiple Libraries and Multiple Types of Tissues

TitleBayesian Analysis of EST Data with Multiple Libraries and Multiple Types of Tissues
Authors
KeywordsDirichlet distribution
Gene expression
Mixture distributions
Multinomial distribution
Shrinkage estimators
Issue Date2006
PublisherAmerican Statistical Association
Citation
American Statistical Association 2006 Joint Statistical Meetings, Seattle, WA, 6-10 August 2006, p. 119 How to Cite?
AbstractESTs (Expressed Sequence Tags) are usually a one-pass sequencing reading of cloned cDNAs derived from a certain tissue. Th e frequency of unique tags among diff erent unbiased cDNA libraries is used to infer the relative expression level of each tag. In this paper, we consider a multinomial model with novel priors of nonlinear Dirichlet distributions for EST data with multiple libraries and/or multiple types of tissues. Th e properties of the priors and the implied posteriors are examined in detail. Gene selection algorithms are developed to detect the co-expression within the same type of tissue and the diff erential expression between diff erent types of tissues. A real EST dataset is used to illustrate the proposed model.
Persistent Identifierhttp://hdl.handle.net/10722/106637

 

DC FieldValueLanguage
dc.contributor.authorYu, Fen_HK
dc.contributor.authorChen, MHen_HK
dc.contributor.authorKuo, Len_HK
dc.contributor.authorHuang, Pen_HK
dc.contributor.authorYang, Wen_HK
dc.date.accessioned2010-09-25T23:24:03Z-
dc.date.available2010-09-25T23:24:03Z-
dc.date.issued2006en_HK
dc.identifier.citationAmerican Statistical Association 2006 Joint Statistical Meetings, Seattle, WA, 6-10 August 2006, p. 119-
dc.identifier.urihttp://hdl.handle.net/10722/106637-
dc.description.abstractESTs (Expressed Sequence Tags) are usually a one-pass sequencing reading of cloned cDNAs derived from a certain tissue. Th e frequency of unique tags among diff erent unbiased cDNA libraries is used to infer the relative expression level of each tag. In this paper, we consider a multinomial model with novel priors of nonlinear Dirichlet distributions for EST data with multiple libraries and/or multiple types of tissues. Th e properties of the priors and the implied posteriors are examined in detail. Gene selection algorithms are developed to detect the co-expression within the same type of tissue and the diff erential expression between diff erent types of tissues. A real EST dataset is used to illustrate the proposed model.-
dc.languageengen_HK
dc.publisherAmerican Statistical Association-
dc.relation.ispartofAmerican Statistical Association Joint Statistical Meetings, JSM 2006en_HK
dc.subjectDirichlet distribution-
dc.subjectGene expression-
dc.subjectMixture distributions-
dc.subjectMultinomial distribution-
dc.subjectShrinkage estimators-
dc.titleBayesian Analysis of EST Data with Multiple Libraries and Multiple Types of Tissuesen_HK
dc.typeConference_Paperen_HK
dc.identifier.emailYang, W: yangwl@hkucc.hku.hken_HK
dc.identifier.authorityYang, W=rp00524en_HK
dc.identifier.hkuros129320en_HK

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats