One of the most powerful tools for stabilizing and boosting yields in any crop is hybrid breeding, which exploits heterosis, the phenomenon by which the offspring of two genetically distinct parents outperform both. In crops like maize and rye, hybrid cultivars have delivered dramatic gains in both yield and yield stability. Faba bean, unfortunately, has never been able to join the hybrid club. Although cytoplasmic male sterility systems, the biological machinery that makes large-scale hybrid seed production possible in other crops, have been identified in faba bean since the 1960s, they have proven too unreliable for commercial use. Breeders have instead turned to synthetic cultivars, which combine at least two inbred parental lines and propagate them together under open pollination. The resulting mixtures capture some of heterosis’s benefits, but at a cost: the best yields typically appear only in the second or third synthetic generation, requiring up to two additional generations of costly field propagation before seed reaches farmers.

A research team working within the BreedPath project, led by scientists at Justus Liebig University Giessen in collaboration with the breeding company NPZ Innovation GmbH, has now tested a strikingly different approach to this problem, one that begins with a camera and a phenomenon first described in 1881. That year, the German botanist Wilhelm Focke noticed that fruits arising from crosses between radish species could be distinguished from those of the parents by their color, an influence of pollen on fruit and seed characteristics that he termed xenien. Today, xenia effects, the measurable impact of paternal pollen on seed traits such as weight, shape, and chemical composition, are documented in crops including maize and pea. In faba bean, earlier studies had reported increases in seed weight and cotyledon cell number in hybrid seeds. The German team reasoned that if hybrid faba bean seeds carry even subtle physical or chemical signatures of their hybridity, hyperspectral imaging might be able to detect them, allowing breeders to sort hybrid seeds from self-pollinated ones without a single DNA test.

The technical setup was ambitious. The researchers assembled eighteen biparental synthetic Syn-1 combinations of spring faba bean, each produced by open-pollinating two nearly homozygous parental lines under isolation tents stocked with bumblebees to guarantee outcrossing. Each Syn-1 lot therefore contained a mixture of F1 hybrid seeds from cross-pollinations and homozygous inbred seeds from self-pollinations. Seeds from each combination, along with pure seeds of the parental lines, were placed into 24-well tissue culture trays and imaged with two hyperspectral line-scanning sensors covering an extraordinary range from 400 to 2500 nanometers, spanning the visible, near-infrared, and short-wave infrared regions. A 1000-watt halogen spotlight provided illumination, and every image was calibrated against a Zenith Polymer diffuser of known reflectance, with dark-current correction applied pixel by pixel. Segmentation algorithms then isolated each individual seed from the tray images, extracting mean reflectance, reflectance variability, and seed size for every single seed in the experiment.

To know the truth about each seed’s genetic status, the team grew all the imaged seeds and genotyped them. Using SNP chip data from the parental lines, they designed Kompetitive Allele Specific PCR markers that could classify each offspring as homozygous, corresponding to one of the two parents, or heterozygous, meaning a true F1 hybrid. This ground-truthing revealed that the proportion of hybrid seeds in the Syn-1 lots ranged from 13.5 to 26.9 percent, with a mean of 17.9 percent, reflecting the modest natural outcrossing rates of the material. With spectral data in hand and genetic labels attached, the researchers deployed five machine learning algorithms: k-nearest neighbors, naive Bayes, C5.0 decision trees, random forest, and support vector machines. Spectral curves were denoised with Savitzky-Golay filtering, hyperparameters were tuned by Bayesian optimization, and performance was assessed through fivefold cross-validation repeated ten times, with accuracy as the target metric for parental classification and the F1 score, a balance of precision and recall, for hybrid identification.

The first question was whether the algorithm could tell the two parental lines of each cross apart from their spectra alone. In the initial global scenario, with 96 seeds per combination, the answer was a qualified yes. The single best combination, a random forest classifier fed with spectral data plus seed weight, size, and cross-identity information, achieved a mean accuracy of 90.5 percent across the eighteen crosses, with individual runs reaching 95.1 percent. Notably, the weight and size data contributed little; the spectral signatures alone carried most of the discriminating power. When the team then resampled two selected combinations with 300 additional seeds each, performance climbed further still, reaching a mean accuracy of 95.6 percent for one combination and an impressive 98.9 percent for the other, where random forest correctly assigned nearly every seed to its parental line. These figures place faba bean alongside wheat, maize, okra, and loofah, where spectral variety identification has previously achieved accuracies of up to 98 percent.

Identifying the hybrid seeds themselves proved far harder. When all seeds in a Syn-1 lot, hybrids and inbreds alike, were thrown into the classification task, the best F1 score in the global scenario was a meager 19.7 percent. Principal component analysis offered an explanation: the spectra of F1 seeds generally fell within the clusters formed by their parental lines, a pattern consistent with strong maternal effects on seed spectral properties, which encompass seed color, known to be maternally inherited in faba bean. In other words, the hybrid seeds looked, spectrally, like their mothers. Yet the picture was not entirely hopeless. When the eighteen crosses were ranked by the spectral distance between their parental components, the more divergent crosses yielded consistently better predictions, echoing earlier findings that xenia effects are more pronounced between genetically distant parents. And in the enlarged second-stage experiments, F1 scores rose to around 38 to 40 percent, above the 33 percent expected from random assignment among three seed classes.

The most intriguing result emerged when the researchers exploited the maternal clustering directly. For one Syn-1 combination whose parental components could be cleanly separated in a principal component analysis, the team assigned each hybrid seed to the parental cluster it resembled, on the hypothesis that maternal effects pulled hybrids toward their seed parent. Analyzing the two parental components separately, they found that for one parent the decision tree algorithm reached an F1 score of 62.2 percent with 64.8 percent accuracy, and for the other, naive Bayes achieved an F1 score of 60.4 percent with 59.0 percent accuracy, both well above the 50 percent random-assignment threshold. A wavelength-specific strategy, restricting the analysis to spectral regions where hybrid seeds deviated from the midpoint between their parents, also nudged performance upward. The authors suggest that future gains could come from hybrid convolutional neural networks that combine spectral reflectance with structural and textural seed features, an approach that has already outperformed single-feature models in wheat variety identification.

The practical implications, while modest on the surface, could be meaningful for breeders. With outcrossing rates in the studied material between roughly 14 and 20 percent, an F1 score of 39 percent would translate into a sorted seed lot containing around 39 percent hybrid seed, up from under 20 percent naturally. Given that heterosis for midparent yield in faba bean has been estimated at 33 to 51 percent, even a partially enriched synthetic cultivar could be expected to yield more than a conventional one. Better still, if hybrid-enriched Syn-1 seed could match the performance that normally requires propagation through Syn-2 or Syn-3 generations, breeders could save an entire generation of expensive seed multiplication. The same spectral pipeline might also serve as a screening tool to identify parental lines with naturally high outcrossing rates, or to replace costly DNA marker checks for confirming the hybridity of offspring from controlled crosses. The method has not yet produced a pure hybrid cultivar, and the authors caution that their results, obtained exclusively from breeding material, still need validation across the remarkable diversity of faba bean germplasm, which ranges from small-seed Minor types to large-seed Major types in colors spanning violet, red, green, beige, brown, black, and spotted. But the study demonstrates, for the first time in a legume, that a camera, a lamp, and a machine learning model can peer into a pile of seeds and begin to pick out the hybrids hiding among them, a small optical trick that could help bring the yield benefits of hybrid breeding to one of the world’s most sustainable protein crops.

Subject of Research: Using hyperspectral imaging and machine learning to differentiate parental and F1 hybrid seeds in faba bean synthetic cultivars

Article Title: Progeny Differentiation in Faba Bean Using Hyperspectral Images and Machine Learning

Article References: Schlichtermann, R.-H., Warnemünde, S., Tietgen, H., Welna, G., Stahl, A., Wittkop, B., & Snowdon, R. J. (2026). Progeny Differentiation in Faba Bean Using Hyperspectral Images and Machine Learning. Plant Direct, 10(10), Article e70193. https://doi.org/10.1002/pld3.70193

Image Credits: AI Generated

DOI: 10.1002/pld3.70193

Keywords: faba bean, hyperspectral imaging, machine learning, xenia effect, hybrid breeding, synthetic cultivars, heterosis, seed sorting, near-infrared spectroscopy, plant breeding, Vicia faba, phenomics

The stakes are enormous. Wheat feeds roughly two billion people daily and covers more than 800 million hectares worldwide, yet global warming threatens to raise growing-season temperatures by one to four degrees Celsius, translating into yield losses of four to six percent. The most damaging heat often arrives during flowering and grain filling, the so-called terminal phase, when high temperatures disrupt photosynthesis, impair pollen viability, and starve developing grains of the sugars they need to plump up. The result is shriveled, lightweight grain and diminished harvests. Heat also triggers the accumulation of reactive oxygen species inside plant tissues, compounds that damage cells unless neutralized by the antioxidant enzymes that distinguish tolerant plants from sensitive ones.

Traditionally, breeders have judged heat tolerance using indices built entirely on yield comparisons between stressed and unstressed plots, measures with names like the stress tolerance index and the yield stability index. The Indian team, led by Rounak Kumar and Deepak Kumar Baranwal, argued that these yield-only metrics miss a richer source of information: the morphology of the grain. Digital image analysis, an emerging phenomics technique in which seeds are arranged on a flat surface and photographed for software to measure, can rapidly quantify size, volume, shape, and color. The researchers reasoned that if specific grain dimensions held steady under heat in tolerant lines but collapsed in sensitive ones, those measurements could become selection criteria in their own right.

The experiment began in the 2021-22 season with 98 elite wheat lines drawn from international nurseries of the International Maize and Wheat Improvement Center, known as CIMMYT, plus two Indian check varieties. The lines were sown twice at the university’s research farm in Sabour, in the subtropical humid east of India: once at the optimal time in late November, and once a month later at the end of December, a widely used trick to force the crop to flower and fill grain under rising spring temperatures. Using a digital grain analyzer, the team recorded nine grain parameters for every genotype, from grain area and perimeter to maximum and minimum grain diameter and roundness.

The heat treatment left a clear statistical fingerprint. Yield per plot fell by an average of 21.91 percent under late sowing, while the grain dimensions shrank more modestly: grain area dropped 10.59 percent, grain width 7.95 percent, average grain diameter 5.44 percent, grain perimeter 4.61 percent, and grain length 3.81 percent. Yet the averages concealed remarkable genetic variation. One entry lost nearly half its plot yield under heat, while another actually gained more than 26 percent. Individual lines bucked the trend on nearly every trait, with some maintaining or even enlarging their grains under stress, evidence that tolerance is woven into the genetics of certain accessions rather than being a universal property of the crop.

Correlation and regression analyses then connected grain geometry to yield performance. Under both timely and late sowing, plot yield correlated positively and significantly with grain area, grain width, minimum and average grain diameter, and grain perimeter, with coefficients ranging from about 0.22 to 0.43. Regression confirmed that grain area, diameter, and width exerted positive effects on yield under heat stress. Interestingly, the shape ratios, roundness and perimeter ratio, stayed comparatively stable even as absolute size shrank, suggesting that heat compresses grain size without distorting grain form. In practical terms, a breeder who selects for plump, large-area grains under late-sown stress is indirectly selecting for yield resilience.

The study then narrowed its focus. Twenty shortlisted entries, chosen using the grain data together with published heat susceptibility indices for grain filling duration, thousand-grain weight, and canopy greenness, plus disease screening and molecular marker surveys, went into a preliminary yield trial in the 2022-23 season. Unusually heavy March rainfall that year, more than 45 millimeters during peak grain filling, complicated grain quality, so the team advanced six entries to a final three-environment trial in 2023-24: optimum sowing in November, late sowing in early December, and extended late sowing at the winter solstice. Yield declined stepwise across the three environments, from 701 grams per plot under optimum sowing to 652.86 grams under late sowing and 636.93 grams under extended late sowing, with thousand-grain weight falling to 35.51 grams in the harshest treatment.

The final-year correlations reinforced the earlier pattern. Across all three environments combined, most traits correlated positively and significantly with plot yield, and grain area in particular tracked yield closely, with regression coefficients of 0.68 under optimum sowing, 0.72 under late sowing, and 0.69 under extended late sowing. Thousand-grain weight was an even stronger predictor, reaching a coefficient of 0.9 under the most extreme heat. Two entries stood out as donor candidates: entry 46, carrying the identifier BRW3989, and entry 92, BRW3992, both of which combined low heat susceptibility for yield and grain weight with promising canopy greenness scores, resistance to spot blotch disease, and favorable morphology. Both have been entered in India’s national genetic stock nursery for 2024-25, making them available to wheat breeders across the country, and segregating families derived from these donors are already under development.

What makes the findings compelling is their mechanistic plausibility. High temperatures suppress starch synthesis and sucrose metabolism in the developing grain, breaking the source-to-sink pipeline that fills kernels. Some genotypes appear to defend that pipeline by maintaining sucrose flux through sucrose synthase enzymes, stabilizing the energy supply to developing grain tissues even as temperatures rise. Grain size, in this light, is not merely a passive outcome of stress but an integrated readout of how well a plant’s vascular plumbing, photosynthetic apparatus, and enzymatic machinery have weathered the heat. A large grain area under late sowing implies that the grain-filling period stayed long enough and the sugar supply steady enough to build a full kernel.

The authors propose that grain area, grain diameter, and grain width be promoted as selection criteria for screening terminal heat-tolerant wheat, offering an alternative or complement to the classic yield-based indices. Because image-based grain analysis is fast and inexpensive once the equipment is in place, it could let breeding programs cull sensitive lines early, before committing scarce field space to multi-year yield trials. The study also carries a broader warning: with roughly six percent of yield lost for every one degree Celsius of warming during grain filling, and with the extended late-sown environment imposing the steepest penalties of all, the window for deploying heat-resilient cultivars is narrowing. For a crop that anchors food security across South Asia and beyond, the geometry of a single grain may prove to be one of the most informative signals breeders can measure.

Subject of Research: Terminal heat stress effects on yield and grain parameters in wheat genotypes

Article Title: Assessment of terminal heat stress on yield and grain parameter in wheat

Article References: Kumar, R., Kumar, M., Singh, P., Pal, A., Vishwakarma, M. K., & Baranwal, D. K. (2026). Assessment of terminal heat stress on yield and grain parameter in wheat. Discover Plants, 3(1), Article 403. https://doi.org/10.1007/s44372-026-00877-6

Image Credits: AI Generated

DOI: 10.1007/s44372-026-00877-6

Keywords: wheat, terminal heat stress, grain parameters, plant breeding, climate change, yield, phenomics, grain area, heat tolerance, CIMMYT, crop science, food security