Mapping genes with longitudinal phenotypes via Bayesian posterior probabilities

Anthony Musolf, Alejandro Q. Nato, Douglas Londono, Lisheng Zhou, Tara C. Matise, Derek Gordon

Research output: Contribution to journalArticlepeer-review

6 Scopus citations

Abstract

Most association studies focus on disease risk, with less attention paid to disease progression or severity. These phenotypes require longitudinal data. This paper presents a new method for analyzing longitudinal data to map genes in both population-based and family-based studies. Using simulated systolic blood pressure measurements obtained from Genetic Analysis Workshop 18, we cluster the phenotype data into trajectory subgroups. We then use the Bayesian posterior probability of being in the high subgroup as a quantitative trait in an association analysis with genotype data. This method maintains high power (>80%) in locating genes known to affect the simulated phenotype for most specified significance levels (α). We believe that this method can be useful to aid in the discovery of genes that affect severity or progression of disease.

Original languageEnglish (US)
Article numberS81
JournalBMC Proceedings
Volume8
DOIs
StatePublished - Jun 17 2014

All Science Journal Classification (ASJC) codes

  • Biochemistry, Genetics and Molecular Biology(all)

Fingerprint Dive into the research topics of 'Mapping genes with longitudinal phenotypes via Bayesian posterior probabilities'. Together they form a unique fingerprint.

Cite this