How to estimate heritability: a guide for genetic epidemiologists

Area of Science:

• Genetic Epidemiology
• Quantitative Genetics
• Genomics

Background:

• Heritability estimation traditionally relied on family-based designs like twin studies.
• Molecular genomics advancements enable estimation using large genotyped samples, including unrelated individuals.

Purpose of the Study:

• To provide a comprehensive overview of common heritability estimation methods.
• To guide readers in understanding and selecting appropriate methods based on study design and sample characteristics.

Main Methods:

• Overview of family-based designs (twin, family studies).
• Description of genomic designs for unrelated individuals (LD Score Regression, GREML).
• Explanation of family-based genomic designs (sibling regression, GREML-kinship, trio-GWAS, relatedness disequilibrium regression).

Main Results:

• Detailed explanation of how heritability is estimated for each method.
• Discussion of the assumptions underlying each method and implications of violations.
• Comparison of benefits and limitations of using related versus unrelated individuals.

Conclusions:

• Understanding the assumptions and limitations of each method is crucial for accurate heritability estimation.
• The choice of method depends on the specific research question, sample availability, and genetic architecture of the trait.
• This guide aids researchers in selecting the most appropriate heritability estimation strategy.