Statistics > Applications
[Submitted on 14 Mar 2012 (this version), latest version 25 Oct 2012 (v3)]
Title:An efficient approach to simultaneous SNP selection: A case study on GAW17 data
View PDFAbstract:Motivation: Identification of causal SNPs in most genome wide association studies relies on approaches that consider each SNP individually. Increasingly, modern regression methods are employed for SNP selection that consider all markers simultaneously and thus take account of dependencies among SNPs. A promising multivariate strategy for prioritizing biomarkers are CAR scores but so far their application has been restricted to moderately sized data sets.
Results: For estimating CAR scores from high-dimensional data we introduce a novel computationally efficient procedure. Subsequently, we conduct a comprehensive comparison study comprising five advanced regression approaches (boosting, lasso, NEG, MCP, and CAR score regression) and a univariate approach (marginal correlation) to find true causal SNPs in data published by the GAW17 consortium. Using this algorithm we demonstrate that SNP rankings based on CAR scores consistently outperform all competing approaches, both uni- and multivariate.
Availability: R code to replicate the complete analysis is available from this http URL
Submission history
From: Korbinian Strimmer [view email][v1] Wed, 14 Mar 2012 13:42:11 UTC (436 KB)
[v2] Fri, 1 Jun 2012 14:11:53 UTC (434 KB)
[v3] Thu, 25 Oct 2012 21:51:38 UTC (436 KB)
Current browse context:
stat.AP
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.