# Transcriptomic prediction of breeding values in loblolly pine

## TL;DR

This study explores using gene expression data to predict genetic values in loblolly pine, improving breeding predictions without needing detailed family records.

## Contribution

The study demonstrates that combining gene expression and SNP data improves predictive accuracy for genetic values in forest trees.

## Key findings

- Specific transcript sets improved predictive power compared to using all transcripts or SNPs.
- Models using both transcript levels and SNPs had higher accuracy than models using only one data type.
- The approach does not require known pedigree relationships, enabling broader use in natural populations.

## Abstract

Phenotypic variation in forest trees can be partitioned into subsets controlled by genetic variation and by environmental factors, and heritability expressed as the proportion of total phenotypic variation attributed to genetic variation. Applied tree breeding programs can use matrices of relationships, based either on recorded pedigrees in structured breeding populations or on genotypes of molecular genetic markers, to model genetic covariation among related individuals and predict genetic values for individuals for whom no phenotypic measurements are available. This study tests the hypothesis that genetic covariation among individuals of similar genetic value will be reflected in shared patterns of gene expression or shared sequence variation in expressed genes. We collected gene expression data by high-throughput sequencing of RNA isolated from pooled seedlings from parents of known genetic value, and compared alternative approaches to data analysis to test this hypothesis. Selection of specific sets of transcripts increased the predictive power of models over that observed using all transcripts or SNPs. Models using information of both transcript levels and SNP variation showed increased predictive accuracy relative to models using only SNPs or transcript levels. Known pedigree relationships are not required for this approach to modeling genetic variation, so it has potential to allow broader application of genetic covariance modeling to natural populations of forest trees.