Genomic Prediction Shows Promise for Breeding Tetraploid Pot Carnation

Carnation (Dianthus caryophyllus L.) ranks among the most commercially important ornamental crops in the world, yet genomic selection has not been widely adopted as a breeding tool in this species. A new study published in BMC Plant Biology examines the genetic architecture of key ornamental traits in a tetraploid pot carnation diversity panel, combining genome-wide association studies (GWAS) with genomic prediction to evaluate how effectively DNA markers can predict trait performance in this panel. The research team, led by Hugo H. Tavera and colleagues, focused on eight traits relevant to growers and consumers of pot carnations: flowering earliness, flower filling, flower count, flower size, branching, plant habit, plant height, and vigour.

To handle this complexity, the researchers compared three different ways of encoding SNP marker data for genomic prediction. The first was a pseudo-diploid encoding. The second was a tetraploid additive dosage encoding, which accounts for the number of copies of each allele.

The third incorporated both additive and non-additive genetic effects. The study assessed all three encodings across the trait set. The GWAS portion of the study identified potential genomic regions of interest associated with seven of the eight traits examined, extending knowledge of where in the carnation genome associations with ornamental quality traits may reside.

Genomic prediction delivered moderate to high accuracy across traits and encodings. Prediction accuracy ranged on average from 0.54 to 0.82 for the flowering-related traits of earliness, flower filling, flower count, and flower size. Plant architecture traits, including branching, habit, plant height, and vigour, showed lower accuracies, ranging from 0.39 to 0.48 on average.

The team also tested a hybrid strategy known as GWAS-assisted genomic prediction, in which the top-scoring markers from the association analysis were fitted as fixed effects within the prediction model. In practice, the gains were modest: accuracy improved by up to 0.03 for three of the eight traits, namely earliness, flower filling, and branching, while the remaining traits showed no improvement under any of the three marker encodings. These results carry practical implications for carnation breeding programs.

The reported accuracies indicate the potential of genomic selection as a breeding tool for tetraploid pot carnation, a crop in which the approach has yet to be widely adopted. Because the study was conducted on a single diversity panel, the reported accuracies reflect this specific population and trait set, and performance could differ in other germplasm or breeding contexts. Nevertheless, the authors conclude that the findings extend knowledge of the genetic basis of key ornamental traits and attest to the potential of genomic selection for breeding tetraploid pot carnation.

Subject of Research: Genome-wide association studies and genomic prediction of ornamental traits in tetraploid pot carnation

Article Title: Genome-wide association studies and genomic prediction for key ornamental traits in tetraploid pot carnation

Article References: Tavera, H. H., Losert, D., Boxriker, M., Boehm, R., Kienbaum, L., Dutta, S., & Zenke-Philippi, C. (2026). Genome-wide association studies and genomic prediction for key ornamental traits in tetraploid pot carnation. BMC Plant Biology. https://doi.org/10.1186/s12870-026-10279-1

Image Credits: AI Generated

DOI: 10.1186/s12870-026-10279-1

Keywords: carnation, genomic selection, GWAS, tetraploid, ornamental breeding, SNP markers, flowering traits, plant architecture, Dianthus caryophyllus, prediction accuracy, Genome-wide, association