A simple nonparametric test for linkage with sib-pair censored event time observations

Area of Science:

• Biostatistics
• Statistical Genetics
• Genetic Epidemiology

Background:

• Phenotype and genotype data are often collected from large pedigrees, including multi-generation nuclear families.
• Sibling event times (e.g., disease onset) can be right-censored, meaning the event did not occur by the observation time.

Purpose of the Study:

• To propose a purely nonparametric test for assessing the independence between sibling event time distributions and their genetic sharing.
• To evaluate the relationship between the Haseman-Elston distance measure and mean genetic sharing identical by descent at a genetic marker.

Main Methods:

• Development of a nonparametric statistical test for analyzing censored sibling event time data.
• The test focuses on the Haseman-Elston distance measure and its relationship with genetic sharing (identity by descent) at a marker.
• Application of the test to data from the Genetic Analysis Workshop 12 and validation through simulation studies.

Main Results:

• The proposed nonparametric test is easily implementable for analyzing sibling pair data with censored event times.
• Simulation studies demonstrate the validity and utility of the new statistical method.
• The test effectively assesses the independence of sibling event time distributions from mean genetic sharing.

Conclusions:

• A novel, easily implementable nonparametric test is presented for genetic analysis of sibling event times with censoring.
• This method provides a valuable tool for understanding the genetic basis of traits influenced by time-to-event data.
• The test's validity is supported by simulation, indicating its potential for real-world genetic studies.