Show simple item record

dc.contributor.authorRuffieux, Heleneen
dc.contributor.authorDavison, Anthony Cen
dc.contributor.authorHager, Jörgen
dc.contributor.authorInshaw, Jamieen
dc.contributor.authorFairfax, Benjamin Pen
dc.contributor.authorRichardson, Sylviaen
dc.contributor.authorBottolo, Leonardoen
dc.date.accessioned2020-07-02T23:31:27Z
dc.date.available2020-07-02T23:31:27Z
dc.date.issued2020-06-29en
dc.identifier.issn1932-6157
dc.identifier.urihttps://www.repository.cam.ac.uk/handle/1810/307552
dc.description.abstractWe tackle modelling and inference for variable selection in regression problems with many predictors and many responses. We focus on detecting hotspots, i.e., predictors associated with several responses. Such a task is critical in statistical genetics, as hotspot genetic variants shape the architecture of the genome by controlling the expression of many genes and may initiate decisive functional mechanisms underlying disease endpoints. Existing hierarchical regression approaches designed to model hotspots suffer from two limitations: their discrimination of hotspots is sensitive to the choice of top-level scale parameters for the propensity of predictors to be hotspots, and they do not scale to large predictor and response vectors, e.g., of dimensions $10^3-10^5$ in genetic applications. We address these shortcomings by introducing a flexible hierarchical regression framework that is tailored to the detection of hotspots and scalable to the above dimensions. Our proposal implements a fully Bayesian model for hotspots based on the horseshoe shrinkage prior. Its global-local formulation shrinks noise globally and hence accommodates the highly sparse nature of genetic analyses, while being robust to individual signals, thus leaving the effects of hotspots unshrunk. Inference is carried out using a fast variational algorithm coupled with a novel simulated annealing procedure that allows efficient exploration of multimodal distributions.
dc.publisherInstitute of Mathematical Statistics
dc.rightsPublisher's own licence
dc.rights.uri
dc.subjectstat.APen
dc.subjectstat.APen
dc.titleA global-local approach for detecting hotspots in multiple-response regressionen
dc.typeArticle
prism.endingPage928
prism.issueIdentifier2en
prism.publicationDate2020en
prism.publicationNameAnnals of Applied Statisticsen
prism.startingPage905
prism.volume14en
dc.identifier.doi10.17863/CAM.54642
dcterms.dateAccepted2020-04-12en
rioxxterms.versionofrecord10.1214/20-AOAS1332en
rioxxterms.versionVoR
rioxxterms.licenseref.urihttp://www.rioxx.net/licenses/all-rights-reserveden
rioxxterms.licenseref.startdate2020-06-29en
dc.contributor.orcidRuffieux, Helene [0000-0002-7113-2540]
dc.contributor.orcidRichardson, Sylvia [0000-0003-1998-492X]
dc.contributor.orcidBottolo, Leonardo [0000-0002-6381-2327]
dc.identifier.eissn1941-7330
rioxxterms.typeJournal Article/Reviewen
pubs.funder-project-idAlan Turing Institute (Unknown)


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record