A ripe ear of wheat seems far removed from a genomics laboratory. Yet part of its future may be found in a history that began long before farming. On 25 August 2026, INRAE presented two complementary studies: one reconstructs genomes of extinct ancestors of flowering plants, while the other compares the marks left by ten millennia of domestication in wheat and barley.

The aim is not to revive an ancient cereal or create an “invincible” plant. It is to identify, within enormous genomes, genes and variants that have already accompanied adaptation to different environments. That map can help breeders decide where to look when studying flowering time, yield, stress responses or the way a crop performs in a particular environment.

This matters as climate change shifts seasons and increasingly combines heat with drought. The Intergovernmental Panel on Climate Change, the IPCC, identifies breeding for heat and drought tolerance as an adaptation option. It also warns that changing cultivars or sowing dates cannot fully offset projected losses at higher levels of warming.

What is a plant paleogenome?

A paleogenome is not necessarily DNA recovered from a fossil seed. In the study published online in Molecular Plant in July 2026, researchers compared the genomes of 84 living flowering-plant species. Using similarities, gene order on chromosomes and known duplication events, they reconstructed ten theoretical ancestral genomes, some dating back more than 200 million years.

The method resembles rebuilding an old book from several modern editions. A sentence moved in every edition of one family reveals part of their shared history. A passage that stayed in the same place and retained a similar function over millions of years deserves particular attention. In genomics, conserved gene neighbourhood is called synteny.

The ancestral genomes become reference maps. Researchers can look for related genes that underpin similar traits or processes in different species. The study demonstrates this with genes linked to yield components, flowering and regulation of DNA activity. A question well studied in one crop can therefore guide research in another where less is known.

Why compare wheat and barley?

Wheat and barley have unusually parallel histories. They were among the first cereals domesticated in the Fertile Crescent around 10,000 years ago. Farming communities then carried and grew them across very different latitudes, temperatures and rainfall patterns. At each stage, people retained seed from plants that best met their needs and local conditions.

The Nature Plants study analysed phenotypic and exome data from 672 wheat and 679 barley accessions held in seed banks. Scientists first searched separately for selection signatures in barley, domesticated emmer, durum wheat and bread wheat. They then compared related genes across those groups.

They found more shared signatures than chance would predict. Some were linked to productivity, grain weight, inflorescence architecture or tillering. Others involved stress and defence responses. The team also found marked north–south differences for several genes, notably between populations from the Horn of Africa and the Mediterranean. A shared history has not erased the value of local adaptation.

How evolutionary history could help breeding

Plant breeding usually starts with a practical goal: flowering that avoids the hottest period, better performance under water stress, disease resistance or more stable yield. Yet a useful trait is rarely controlled by one simple genetic switch. It reflects many genes, their regulation and their interaction with soil, weather and farm management.

Comparing crops that encountered similar pressures provides an additional filter. If the same gene, or a closely related one, carries a selection signature in both wheat and barley, it becomes a strong candidate for testing. If a useful variant is known in one cereal but not the other, breeders gain a new lead. The researchers describe this as inter-crop translational research.

INRAE reports that about one hundred variants with potential value emerged from the wheat–barley comparison. The role of three genes previously known in other species was also validated in wheat, relating respectively to yield, flowering time and epigenetic regulation. The essential word is potential. A signal detected in genomic data still has to be tested in fields, across environments and over several years.

Two open tools to narrow the search

The teams released two free resources. AGR, short for Ancestral Genome Reconstruction, compares modern genomes and reconstructs ancestral architectures. OrthoViewer enables users to explore conserved genes across 84 agriculturally important species. Its database includes 1,142 genes whose biological function or agronomic value has been described in scientific literature.

These tools do not choose a variety for an agronomist. They narrow the search space. Instead of examining thousands of genes at random, a team can begin with candidates whose roles or genomic context are conserved across crops. The expected benefit is better-targeted experiments, not an end to experiments.

The approach extends beyond wheat and barley. Researchers are continuing comparisons with rice, maize and sorghum. This may be especially valuable for crops with fewer genomic resources or smaller breeding programmes: robust knowledge from one species can guide work in another, provided the proposed mechanism is tested there.

Why this does not create a “climate-proof” variety

Climate is not a single stress. A plant may cope better with limited water but remain vulnerable to extreme heat during flowering. An early variety may escape late-season stress but have less time to fill its grain. A trait that helps in deep soil may matter less in a shallow field. Disease, nitrogen availability and the quality required by food processors also remain important.

The IPCC makes the same point for European agriculture. Breeding for heat and drought tolerance can improve sustainability, but cultivar and sowing-date changes do not fully compensate for projected losses at high warming levels. Adaptation therefore combines varietal diversity with water-retentive soils, rotations, adjusted sowing dates, careful irrigation where available and local observation.

A statistical association must also be distinguished from agronomic proof. Selection signatures show that a genomic region was probably favoured during history. They do not automatically reveal which variant will perform best in France in 2035. Multi-location trials, comparisons with control varieties and food-quality testing remain necessary before registration and use by farmers.

Seed diversity is a practical insurance policy

These studies highlight the value of seed-bank collections. An accession conserved for decades may carry a variant that becomes useful against an emerging disease, a warmer season or a different rainfall pattern. Conserving diversity is therefore not only about heritage. It preserves options for future research and farming.

That diversity must also be evaluated in real conditions. Modern varieties, landraces and wild relatives are not interchangeable. Each may contribute a useful characteristic, but it must be combined with yield, quality, stability and the needs of growers and food users. Genomics gives a clearer view inside the toolbox; it cannot replace biological time or field knowledge.

What does this mean for farmers and gardeners?

In the short term, nothing changes in a bag of seed. These results are research and pre-breeding resources. Turning them into a commercial variety will require crossing, trials and regulatory evaluation. They are also a reason to be sceptical of claims that one gene offers an immediate answer to climate change.

For farmers, the useful question remains local: which varieties were tested in the region, on which soils, and how did they perform in wet and dry years? Gardeners can diversify varieties and record sowing, flowering and harvest dates. Such notes are not a scientific trial, but they create a small local memory and reduce dependence on one variety or calendar.

The results also reinforce a broader lesson: resilient harvests are often diverse harvests. Plants do not all react in the same way to the same weather event. Across a territory, different crops, varieties and calendars can reduce the risk that everything matures—or fails—at once.

Diverse harvests need flexible local routes

When maturity dates shift or yields become irregular, nearby outlets become more useful. A small surplus of grain, flour, vegetables or seed that may legally be exchanged does not always fit a standardised long supply chain. It may still meet a nearby need, subject to the rules that apply to each product.

Seeed enables individuals and producers to give, sell or swap harvests and local products with people nearby. That connection does not replace plant breeding or cereal supply chains. It can simply help different local outputs complement one another and keep edible surpluses from going unused.

The strongest conclusion from this research is deliberately modest: the genetic past of crops contains useful clues, not a ready-made recipe. Combined with rigorous trials, appropriate farming practices and diversity that is actually grown, those clues can widen the choices available in a less predictable climate.

Sources

Cover image credit

“Barley ears”, photograph by Christoph Strässler, via Wikimedia Commons, licensed under CC BY-SA 2.0. The image is suitable for cropping; no editorial alteration beyond website display.