Multiple Loci Mapping via Modelâfree Variable Selection
2012
Sun, Wei | Li, Lexin
Despite recent flourish of proposals on variable selection, genomeâwide multiple loci mapping remains to be challenging. The majority of existing variable selection methods impose a model, and often the homoscedastic linear model, prior to selection. However, the true association between the phenotypical trait and the genetic markers is rarely known a priori, and the presence of epistatic interactions makes the association more complex than a linear relation. Modelâfree variable selection offers a useful alternative in this context, but the fact that the number of markersâpâoften far exceeds the number of experimental unitsânârenders all the existing modelâfree solutions that requireânâ>âpâinapplicable. In this article, we examine a number of modelâfree variable selection methods for smallânâlargeâpâregressions in the context of genomeâwide multiple loci mapping. We propose and advocate a multivariate groupâwise adaptive penalization solution, which requires no model prespecification and thus works for complex traitâmarker association, and handles one variable at a time so that works forânâ<âp. Effectiveness of the new method is demonstrated through both intensive simulations and a comprehensive real data analysis across 6100 gene expression traits.
Показать больше [+] Меньше [-]Ключевые слова АГРОВОК
Библиографическая информация
Эту запись предоставил National Agricultural Library