Identifying loci affecting trait variability and detecting interactions in genome-wide association studies

Area of Science:

• Genetics
• Statistical Genetics
• Bioinformatics

Background:

• Identifying genetic variants influencing trait variability is crucial for understanding biological mechanisms and interactions.
• Current methods may not fully capture genetic effects on trait variance, especially in non-normal distributions.

Purpose of the Study:

• To propose and implement a novel statistical test for jointly assessing mean and variance effects of genetic variants.
• To develop a specific test for dispersion effects in non-normal trait distributions.
• To increase the power for detecting genetic loci associated with trait variability.

Main Methods:

• A two-degree-of-freedom test implemented within a linear mixed model framework.
• Development of an efficient algorithm and software for the proposed test.
• Application to body mass index (BMI) data from the UK Biobank (British ancestry, n~408,000).

Main Results:

• The proposed approach demonstrated increased power to detect associated genetic loci.
• Novel associations with significant variance effects were identified and replicated.
• These variance effects were not solely attributable to the non-normality of the body mass index trait.
• Suggestive evidence linking leptin levels to body mass index variability was found.

Conclusions:

• The developed statistical test effectively identifies genetic variants impacting trait variability.
• The method enhances the discovery of genetic associations, particularly for complex traits.
• Findings suggest a potential role for leptin in modulating body mass index variability.