Partial genotyping at polymorphic markers can improve heritability estimates in sibling groups

Area of Science:

• Evolutionary Biology
• Quantitative Genetics
• Plant Sciences

Background:

• Accurate heritability (h2) estimates are vital for assessing population adaptive potential and trait evolution.
• Traditional plant heritability studies often use maternal progeny designs, assuming half-sib or unrelated offspring, which can be biased by complex mating systems.

Purpose of the Study:

• To investigate accurate heritability estimation in nonmodel plant species using sibling designs with moderate genotyping.
• To assess the impact of nonrandom mating, inbreeding, and maternal effects on heritability estimates.
• To compare traditional methods with marker-based and animal model approaches.

Main Methods:

• Simulations were conducted to evaluate heritability estimation methods.
• Microsatellite marker data from a subset of offspring were incorporated into maternal progeny designs.
• Methods compared included basic family, adjusted family, and animal models with various relatedness matrices (average, hybrid, pedigree-based).

Main Results:

• Marker-based relatedness methods performed comparably to pedigree-based methods under nonrandom mating (e.g., unequal contributions, selfing).
• Maternal effects required pedigree-based pairwise relatedness for improved accuracy, with father-offspring relatedness being critical.
• The most robust method combined one-generation pedigree and average relatedness coefficients across diverse ecological scenarios.

Conclusions:

• Marker-based relatedness offers a powerful tool for in situ heritability studies in natural plant populations, overcoming limitations of traditional designs.
• Accurate heritability estimation in plants necessitates careful consideration of mating systems and potential maternal effects.
• A hybrid approach integrating pedigree and marker-derived relatedness provides a robust framework for quantitative genetic studies in diverse plant species.