Regularized principal components of heritability

Yixin Fang, Yang Feng, Ming Yuan

Research output: Contribution to journalArticlepeer-review

Abstract

In family studies with multiple continuous phenotypes, heritability can be conveniently evaluated through the so-called principal-component of heredity (PCH, for short; Ott and Rabinowitz in Hum Hered 49:106-111, 1999). Estimation of the PCH, however, is notoriously difficult when entertaining a large collection of phenotypes which naturally arises in dealing with modern genomic data such as those from expression QTL studies. In this paper, we propose a regularized PCH method to specifically address such challenges. We show through both theoretical studies and data examples that the proposed method can accurately assess the heritability of a large collection of phenotypes.

Original languageEnglish (US)
Pages (from-to)455-465
Number of pages11
JournalComputational Statistics
Volume29
Issue number3-4
DOIs
StatePublished - Jun 2014

Keywords

  • Expression quantitative trait loci
  • Family study
  • High dimensional data
  • Linear discriminant analysis
  • Principal components
  • Sparsity

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty
  • Computational Mathematics

Fingerprint

Dive into the research topics of 'Regularized principal components of heritability'. Together they form a unique fingerprint.

Cite this