Comparing Methods for Estimating SNP Effects in Genomic Selection
Author Information
Author(s): Pimentel Eduardo da Cruz Gouveia, Sargolzaei Mehdi, Simianer Henner, Schenkel Flávio Schramm, Liu Zengting, Fries Luiz Alberto, de Queiroz Sandra Aidar
Primary Institution: Georg-August-University of Göttingen
Hypothesis
This research compares the application of iterative and direct solvers for the estimation of SNP effects with potential use on genomic selection.
Conclusion
The iterative solver was found to be faster than direct methods for larger numbers of markers, while the Gentleman-Givens method provided exact solutions in a reasonable time frame.
Supporting Evidence
- The iterative solver was substantially faster than direct methods for larger numbers of markers.
- Use of a direct solver may allow for computing (co)variances of SNP effects.
- Performance of the iterative method varied depending on the level of ill-conditioning of the coefficient matrix.
- Results showed that Gentleman-Givens rotations provided an exact solution within a reasonable time frame.
Takeaway
This study looked at different ways to calculate the effects of genetic markers on traits in animals, finding that one method was faster and better for larger datasets.
Methodology
The study used simulated data and real data from Holstein-Friesian sires, applying one iterative and two direct methods to estimate SNP effects.
Limitations
The performance of the iterative method varied depending on the level of ill-conditioning of the coefficient matrix.
Participant Demographics
3081 Holstein-Friesian sires genotyped for 45181 SNPs.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website