In a major step toward climate-resilient maize, researchers in Germany have created a powerful new genetic resource that reveals how the world’s most important cereal crop can be bred to withstand the heat and drought that increasingly batter European farmland. By tracking the inheritance of millions of DNA variants across nearly 400 inbred maize lines grown in dozens of field trials, a team led by the Technical University of Munich and seed company KWS has mapped 22 regions of the maize genome that shape grain yield under stressful conditions, and has shown that two simple, visually scorable traits—how slowly a plant’s leaves die and how tightly they curl—could help breeders pick out the toughest, highest-yielding lines.
The study, published in Theoretical and Applied Genetics, centers on a so-called MAGIC population: a Multi-parent Advanced Generation Inter-Cross comprising 388 doubled-haploid lines derived from eight founder inbred lines. MAGIC populations are constructed by inter-mating several genetically distinct parents over multiple generations, so that each descendant’s genome becomes a fine-grained mosaic of DNA segments inherited from all eight founders. This design confers distinct statistical advantages over both traditional bi-parental crosses, which capture only limited diversity, and diversity panels, which suffer from uneven population structure that can confound genetic associations. Because the founding genotypes are known exactly and allele frequencies are balanced, MAGIC populations are particularly well suited to a modern, low-cost genotyping strategy known as low-coverage whole-genome sequencing.
The eight founders—B106, B107, F888, FC1890, Lo1056, Lo1270, Lo1290 and PHG83—were deliberately drawn from the DROPS diversity panel, a collection of 244 dent maize hybrids that had previously been evaluated across 29 field experiments spanning nine European sites and one site in Chile. Crucially, the founders performed similarly under irrigated, favorable conditions but diverged sharply under rainfed, stress-prone environments, making them ideal raw material for dissecting drought and heat tolerance. The founders were also chosen to represent major heterotic groups of European dent maize, including Iodent, Lancaster and Non-Stiff Stalk material, while keeping flowering time variation to a maximum of roughly 8 to 11 days to avoid confounding yield differences with maturity differences.
To genotype the population, the researchers sequenced the eight founders at high depth above 50-fold coverage and each of the 388 doubled-haploid lines at an average of 5.7-fold coverage using an Illumina NovaSeq 6000 platform. At such shallow depths, many genomic positions are covered by only one or a few reads, so the team relied on a founder-guided variant calling pipeline: more than 8 million bi-allelic SNPs identified in the founders served as a reference set for calling variants in the progeny. After trimming, error correction, duplicate removal, alignment to the B73 reference genome, filtering of transposable-element regions, and statistical imputation using the HBimpute package, the final dataset contained 2,717,240 high-quality SNPs with no missing data. Comparison against a 600,000-marker SNP array showed an average genotyping error of just 0.08 percent—remarkably close to the 0.04 percent error observed in deep sequencing of the founders.
A key practical outcome of the study is a set of recommendations for how deeply such populations need to be sequenced. By computationally down-sampling high-depth data to coverages ranging from 0.1-fold to 8-fold, the team found that 2-fold coverage represents a turning point: below it, the number of usable SNPs drops sharply, while above it, gains in genome coverage, genotyping rate and imputation accuracy begin to plateau. At 2-fold depth, roughly 75 percent of known SNP loci were retained with less than 50 percent missingness—enough for effective imputation and, given fixed library-preparation costs, the best balance of cost and performance. Even 0.5-fold coverage proved sufficient for identity-by-descent-based mapping in this population, offering a budget option for resource-limited laboratories.
On the phenotyping side, the 388 lines were evaluated between 2020 and 2023 in 21 field trials across seven locations in Germany, Hungary and Italy—seven trials measuring testcross performance with a flint tester line and fourteen trials assessing the lines themselves. The researchers scored ten traits, including grain yield calibrated to 85 percent dry matter, plant and ear heights, male and female flowering times, and three so-called proxy traits long associated with drought response in maize: leaf senescence scored on a 1-to-9 scale four to six weeks after flowering, leaf rolling scored on a hot, dry day, and the anthesis-silking interval, the time lag between pollen shedding and silk emergence. Environmental data spanning a 60-day window around flowering, including precipitation, reference evapotranspiration and maximum temperature, allowed the team to classify individual trials as optimal or suboptimal. In 2021 in Hungary, for example, high temperatures and severe water deficit cut testcross grain yield by 32 percent in the rainfed trial relative to its irrigated counterpart.
The genetic dissection of yield delivered striking results. Genome-wide association analyses identified 22 QTL—quantitative trait loci—for testcross grain yield, jointly explaining 45 percent of the genetic variance. Five of these QTL showed consistent effects across all seven testcross trials, with main effects ranging from 0.27 to 0.42 tonnes per hectare, while the remaining 17 displayed significant QTL-by-trial interactions, meaning their influence on yield depended on the environment. One locus on chromosome 8, qGDY(TC)08A, carried the strongest consistent effect, with favorable alleles contributed by founders PHG83 and Lo1290. Other loci behaved differently under stress versus optimal conditions: a QTL on chromosome 2 flipped the direction of its effect between the two scenarios, and a locus on chromosome 9 exerted its influence almost exclusively in the harsh heat-and-drought environment of Murony in 2021.
The proxy traits told a subtler story. In the stressed rainfed trial at Murony, grain yield correlated significantly with both leaf senescence and leaf rolling—plants whose leaves stayed greener longer and rolled less yielded more, and the top 10 percent of genotypes showed distinctly better stay-green and minimal rolling. The anthesis-silking interval, by contrast, showed no significant correlation with yield, apparently because modern hybrids have already been bred for such short intervals—averaging just 2.3 days under stress in this study—that the trait offers little remaining selection value at the hybrid level. When the researchers fitted bivariate multi-trait statistical models that jointly estimated SNP effects on yield and on each proxy trait, they found that alleles associated with delayed senescence and reduced leaf rolling generally had positive effects on grain yield. Two QTL in particular—qLS(LP)01C for leaf senescence and qLR(LP)03C for leaf rolling—showed consistent effects across trials and substantial influence on testcross yield, marking them as promising targets. Notably, the leaf rolling QTL co-localized with a leaf angle locus identified previously in the maize NAM population, hinting at a shared genetic basis for leaf architecture and drought response.
The team also explored whether these secondary traits could sharpen genomic prediction, the statistical approach in which genome-wide markers are used to estimate the genetic value of breeding lines. Under cross-validation scenarios where leaf senescence and leaf rolling measurements were available for the lines being predicted, multi-trait models modestly improved yield prediction accuracy at the Hungarian stress site, from 0.53 to 0.56 under rainfed conditions and from 0.47 to 0.51 under irrigated conditions, with leaf senescence emerging as the main driver of the gain. The improvements were modest, reflecting the only moderate correlations between the proxy traits and yield itself, but the researchers argue the approach could become genuinely useful if the proxy traits are measured at scale—potentially by drone-based high-throughput phenotyping in compact observation plots.
Beyond yield, the study uncovered loci with clear breeding relevance. A plant height QTL on chromosome 1 with an effect of roughly 11 centimeters lies about 2.2 megabases from Brachytic2, the well-known gene behind short-stature “smart corn” hybrids, and sequencing revealed a potential one-kilobase duplication within the B106 founder’s copy of the gene that may underlie its height-reducing allele. Several candidate genes for leaf senescence identified in earlier work—including a trehalose-6-phosphate synthase and a trihelix transcription factor—fell within the team’s QTL intervals, providing a shortlist for functional follow-up.
The authors suggest their combined findings offer a template for future crop genetics: combine the balanced recombination of MAGIC designs with low-coverage sequencing at around 2-fold depth, deploy both SNP-based and haplotype-based mapping approaches in parallel, and treat stress-related proxy traits not as magic bullets but as scalable secondary signals that, when integrated into multi-trait prediction models, can nudge breeders toward lines better equipped to keep filling kernels when the rain stops and the heat rises. As climate volatility intensifies across the European grain belt, resources like this eight-founder population may prove instrumental in keeping maize fields productive under conditions their ancestors never had to endure.
Subject of Research: Genetic dissection of grain yield and drought-related proxy traits in an eight-founder MAGIC maize population
Subject of Research: Agriculture
Article Title: Genetic dissection of grain yield and correlated proxy traits under suboptimal conditions
Article References: Lin, Y.-C., Urbany, C., Shlykova, A., Hölker, A. C., Ouzunova, M., Presterl, T., Pook, T., Mayer, M., Urzinger, S., & Schön, C.-C. (2026). Genetic dissection of grain yield and correlated proxy traits under suboptimal conditions. Theoretical and Applied Genetics, 139(9), Article 257. https://doi.org/10.1007/s00122-026-05364-w
Image Credits: AI Generated
DOI: 10.1007/s00122-026-05364-w
Keywords: maize, MAGIC population, grain yield, drought tolerance, leaf senescence, leaf rolling, QTL mapping, low-coverage whole-genome sequencing, genomic prediction, doubled haploid lines, heat stress, genotype-by-environment interaction
Cite Scienmag News
APA MLA Chicago
Juliet Wilcox. (September 3, 2026). Scientists uncover genes controlling grain yield in harsh growing conditions. Scienmag. https://scienmag.com/scientists-uncover-genes-controlling-grain-yield-in-harsh-growing-conditions/
Juliet Wilcox. “Scientists uncover genes controlling grain yield in harsh growing conditions.” Scienmag, 3 September 2026, https://scienmag.com/scientists-uncover-genes-controlling-grain-yield-in-harsh-growing-conditions/. Accessed 3 September 2026.
Juliet Wilcox. “Scientists uncover genes controlling grain yield in harsh growing conditions.” Scienmag. September 3, 2026. https://scienmag.com/scientists-uncover-genes-controlling-grain-yield-in-harsh-growing-conditions/
Copy citation Download RIS
Tags: advanced breeding techniques for drought resistancebreeding strategies for climate-adapted cerealClimate-resilient maize geneticsDNA variants and crop resilienceDNA variants in crop improvementdrought and heat stress tolerance in cropsdrought and heat tolerance in cereal cropsEuropean maize cultivation under climate stressgenetic basis of leaf wilting and curling in maizegenetic mapping of grain yieldgenetic markers for drought resiliencegenome regions controlling stress tolerancegenomic regions influencing grain yield under stresshigh-throughput field trials for stress adaptationidentifying yield-related traits in maizeMAGIC population in plant breedingMAGIC populations in plant breedingmaize genetic resource developmentmaize genome mapping for yield traitsmulti-parent advanced generation inter-cross (MAGIC) populationsphenotypic traits for selecting drought-tolerant maizeplant breeding for climate changevisual traits for selecting high-yield lines
