Partial least square regression applied to the QTLMAS 2010 dataset
2011

Using Partial Least Square Regression to Identify Genetic Traits

publication Evidence: moderate

Author Information

Author(s): Coster Albart, Calus Mario P L

Primary Institution: Animal Breeding and Genomics Centre, Wageningen University

Hypothesis

Can partial least square regression effectively identify genomic regions affecting traits and estimate breeding values?

Conclusion

The study demonstrated that partial least square regression is a viable method for analyzing quantitative trait loci and estimating breeding values.

Supporting Evidence

  • The study identified 25 QTL for a continuous trait and 22 for a discrete trait.
  • The accuracies of estimated breeding values ranged from 0.56 to 0.92.
  • Pleiotropic QTL were found on chromosome 1.

Takeaway

This study used a special math method to find important genes that affect traits in animals and to predict how good they are for breeding.

Methodology

The study used partial least square regression to analyze genetic data and estimate breeding values.

Potential Biases

There is a risk of biased regression coefficients for QTL in linkage disequilibrium with other QTL.

Limitations

The method may lead to false detections of QTL and does not provide a quantitative test statistic for QTL presence.

Digital Object Identifier (DOI)

10.1186/1753-6561-5-S3-S7

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication