Integrating mRNA transcripts and genomic information into genomic prediction.

Area of Science:

• Quantitative Genetics
• Genomics
• Bioinformatics

Background:

• Genomic prediction is crucial for livestock, crops, and human disease risk assessment.
• Classical methods struggle to incorporate biological prior information, like genetic regulation mechanisms.
• There is a need for advanced genomic prediction methods that leverage biological insights.

Purpose of the Study:

• To introduce and evaluate a novel approach for genomic prediction using mRNA transcript information.
• To assess the impact of integrating mRNA transcript data on complex trait phenotype prediction accuracy.
• To compare the new method's performance against traditional genomic best linear unbiased prediction (GBLUP).

Main Methods:

• Utilized a *Drosophila* population, a standard model in quantitative genetics research.
• Integrated mRNA transcript data into genomic prediction models.
• Compared prediction accuracy of the novel method with GBLUP for various traits.

Main Results:

• Integrating mRNA transcript data significantly improved genomic prediction accuracy for specific traits in *Drosophila*.
• Observed improvements include a rise in accuracy for olfactory response to dCarvone (0.256 to 0.274) and cafe (0.355 to 0.401) in males.
• Accuracy for survival_paraquat in males improved (0.101 to 0.138), and olfactory response to 1hexanol in females (0.147 to 0.210).

Conclusions:

• Integrating mRNA transcripts substantially enhances genomic prediction accuracy for certain complex traits, with improvements ranging from 7% to 43%.
• The novel approach offers a significant advancement over GBLUP for specific trait predictions.
• Incorporating interaction effects alongside mRNA transcript integration can further boost prediction accuracy for some traits.