Statistical properties of affected sib-pair linkage tests

Chih Chieh Wu, Christopher I. Amos

Research output: Contribution to journalArticlepeer-review

3 Citations (Scopus)


Genetic linkage analysis is a powerful tool for the identification of disease susceptibility loci. Among the most commonly applied genetic linkage strategies are affected sib-pair tests, but the statistical properties of these tests have not been well characterized. Here, we present a study of the distribution of affected sib-pair tests comparing the type I error rate and the power of the mean test and the proportion test, which are the most commonly used, along with a novel exact test. In contrast to existing literature, our findings showed that the mean and proportion tests have inflated type I error rates, especially when used with small samples. We developed and applied corrections to the tests which provide an excellent adjustment to the type I error rate for both small and large samples. We also developed a novel approach to identify the areas of higher power for the mean test versus the proportion test, providing a wider and simpler comparison with fewer assumptions about parameter values than existing approaches require.

Original languageEnglish
Pages (from-to)153-162
Number of pages10
JournalHuman Heredity
Issue number4
Publication statusPublished - 2003

All Science Journal Classification (ASJC) codes

  • Genetics
  • Genetics(clinical)

Fingerprint Dive into the research topics of 'Statistical properties of affected sib-pair linkage tests'. Together they form a unique fingerprint.

Cite this