Estimating Genetic Values with Hierarchical Likelihood
Author Information
Author(s): Shen Xia, Rönnegård Lars, Carlborg Örjan
Primary Institution: The Linnaeus Centre for Bioinformatics, Uppsala University
Hypothesis
Can hierarchical likelihood improve the estimation of genetic values using genome-wide dense marker maps?
Conclusion
Hierarchical likelihood allows for efficient estimation of marker-specific variances and accurate localization of major QTL.
Supporting Evidence
- The study achieved a Pearson correlation of 0.60 for quantitative traits and 0.72 for binary traits in young individuals without phenotypic records.
- Hierarchical likelihood was shown to be computationally faster than Bayesian methods.
- The method accurately localized major QTL with high precision.
Takeaway
This study shows a new way to find important genes in animals by using a smart math method that works faster than older methods.
Methodology
The study used double hierarchical generalized linear models to analyze a simulated dataset for QTL mapping and GEBV estimation.
Limitations
The study did not consider joint modeling of both quantitative and binary traits.
Participant Demographics
The dataset included 3226 individuals across 5 generations.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website