Study of 206 Riesling clones finds epigenomic data better explain year-to-year shifts

Researchers studying Riesling grapevines have reported evidence that genetic and epigenetic data can capture different parts of the way vineyard performance shifts from year to year, a finding that could eventually help breeders and growers select clones that perform more consistently under changing conditions.

The work appeared in a preprint posted Friday, November 6, on bioRxiv and has not been peer reviewed. The study was led by scientists at Hochschule Geisenheim University in Germany, with collaborators in Australia and at the University of Hohenheim.

The researchers analyzed 206 Riesling clones over nine growing seasons and examined traits tied to both production and juice quality. Their focus was genotype-by-environment interaction, a long-standing problem in perennial crops such as grapevines. In practical terms, that interaction means a clone that performs well in one year may not rank the same way in another year because weather and other environmental conditions change.

The team used a method called UMR-seq, which allowed them to generate paired genomic and epigenomic profiles from the same sample library. The genomic data tracked DNA sequence differences, including SNPs. The epigenomic data focused on unmethylated regions, a DNA methylation signal that can reflect changes in gene regulation without altering the genetic code itself.

Using those paired datasets, the researchers tested how much each layer of information explained differences among clones and differences in how clones responded across years. According to the preprint, when the statistical models did not explicitly account for genotype-by-environment interaction, environmental and residual effects made up the largest share of variation in the measured traits, while the main clonal effect was relatively modest. In that setting, genomic relationships explained most of the clonal variance in four of the five traits studied, and adding an epigenomic relationship matrix made little difference.

The picture changed when the researchers modeled year-specific interaction effects more directly. In those analyses, the epigenomic layer accounted for a substantial share of genotype-by-environment interaction variance and contributed more than the genomic layer in four of the five traits, the authors wrote. That suggests the epigenome may be especially relevant for understanding why the same clone can behave differently across seasons.

The study also found that the two data layers were not simply measuring the same thing in different ways. The authors reported that genomic and epigenomic relationships captured different patterns of clone re-ranking across years, meaning each one identified distinct shifts in relative performance as conditions changed from season to season. When the team mapped the strongest estimated effects onto genomic windows, the largest genomic and epigenomic signals tended to fall in different regions, with overlap no greater than would be expected by chance.

The findings do not show that epigenetic markers alone can predict stable performance in vineyards, and the preprint does not present a direct tool for clone selection. Still, the results point to a possible way to improve selection in clonally propagated crops by combining conventional genomic information with epigenomic data, especially where year-to-year variability is a major concern.

That possibility matters beyond plant breeding theory. In wine production, seasonal swings can affect both grape yield and must composition, two factors that shape supply and quality. If future work confirms that epigenomic data helps identify clones that hold steadier performance across variable conditions, breeders and nurseries could eventually use that information to guide clonal choices for vineyards facing unstable weather. Any such application remains tentative and would need validation outside this preprint.

The paper places its results in the broader challenge of breeding perennial crops. Because grapevines are propagated clonally and remain in the field for many years, selection decisions can be costly and slow to reverse. A clone that looks promising in a limited trial may not deliver the same result over time if its performance is highly sensitive to annual conditions. For that reason, understanding what drives stability across years is a central issue for viticulture and grape breeding.

The authors said their results provide a basis for investigating the molecular drivers of genotype-by-environment interaction and for testing whether genomic and epigenomic information can improve the selection of clones for variable environments. The study does not settle how durable epigenomic signals are over time or how broadly the findings apply beyond Riesling, one site, or the traits examined. It also leaves open whether similar patterns would appear in other grape varieties or other perennial crops.

Still, the work adds to growing interest in whether epigenetic variation can help explain agricultural traits that standard genetic markers do not fully capture. In this case, the main signal from the preprint is not that epigenomics replaced genomics, but that the two appeared to describe different components of performance. Genomic relationships were more informative for clonal main effects in most traits, while epigenomic relationships were more informative for a large part of the year-to-year interaction pattern.

Because the paper remains a preprint, the methods and interpretation have yet to be tested through peer review. Even so, the dataset is unusually broad for a vineyard clone study, covering 206 Riesling clones across nine seasons, and it offers an early look at how paired genomic and epigenomic measurements might be used to study instability in grape production over time.