The research, conducted during the 2022/2023 main cropping season, took place across two contrasting districts in Eastern Amhara: Dawa Cheffa in the South Wollo Zone and Ambasel in the Oromia Special Zone. These locations represent the major mung bean-growing agro-ecologies of the region, sitting at altitudes of 1,400 to 1,680 meters above sea level on vertisol soils. Cheffa receives roughly 850 millimeters of annual rainfall while Jari receives about 1,204.6 millimeters, providing a natural gradient of moisture conditions under which to test varietal performance. The study employed a participatory varietal selection approach, a methodology that deliberately places farmers at the center of the evaluation process rather than treating them as passive recipients of researcher decisions.

Four mung bean varieties entered the trials: three improved types released by Ethiopian agricultural research centers, namely Rasa (also designated N-26), NVL-01, and Boreda, alongside Shewa Robit, a widely grown local variety that served as the farmers’ familiar benchmark. The experimental design followed what researchers call the mother and baby trial approach. Mother trials, managed under strict scientific protocols at research sites, used a randomized complete block design with four replications, planting each variety in five-row plots of three meters with 0.4-meter row spacing and 5-centimeter plant spacing. Baby trials, conducted on farmers’ own fields with each farmer serving as a replication, used larger 25-square-meter plots of five by five meters. This dual structure allowed researchers to capture both statistically rigorous agronomic data and the messier, more nuanced realities of how varieties perform under actual farm conditions.

The combined analysis of variance revealed statistically significant differences among the varieties for every parameter measured, from days to flowering through days to maturity, plant height, pods per plant, seeds per pod, hundred seed weight, and ultimately grain yield. Rasa emerged as the clear agronomic leader, producing a mean seed yield of 1,731 kilograms per hectare across both locations. Boreda followed closely at 1,690 kilograms per hectare, a difference that, while numerically apparent, did not reach statistical significance. Both varieties substantially outperformed NVL-01, which yielded 1,225 kilograms per hectare, and the local check Shewa Robit, which produced only 1,020 kilograms per hectare. The yield gap between the best improved variety and the local landrace represents a potential 70 percent increase simply from switching varieties, a margin that could fundamentally alter household economics in the region.

Beyond raw yield, the physiological differences between varieties told an important story. Rasa and Boreda both achieved a hundred seed weight of 5.6 grams, significantly heavier than NVL-01 at 5.3 grams and Shewa Robit at just 3.7 grams. Heavier seeds generally correlate with better market acceptance and higher prices. Rasa also produced the most pods per plant at 15.12, followed by Boreda at 14.80, NVL-01 at 11.95, and Shewa Robit at 10.15. Seeds per pod followed the same hierarchy, with Rasa producing 11.70 seeds per pod compared to Shewa Robit’s 9.28. Plant height, which matters to farmers because taller plants produce more straw for animal feed, ranged from 55.27 centimeters in Rasa down to 51.32 centimeters in Shewa Robit, a statistically significant difference that farmers noted during their field evaluations.

Disease pressure emerged as one of the most decisive factors separating the varieties. Cercospora leaf spot and anthracnose, two of the most destructive mung bean pathogens in the region, ravaged the local Shewa Robit variety and, to a lesser extent, NVL-01. Shewa Robit recorded an anthracnose score of 3.5 on a five-point severity scale, the worst of any variety tested, while Rasa scored just 1.875, indicating substantially better tolerance. For farmers who have watched entire seasons decimated by disease, this difference alone justified switching. The study’s authors note that these diseases, along with yellow mosaic virus and dry root rot, continue to severely limit mung bean productivity across the region, making genetic resistance a top priority in any breeding or selection program.

The participatory component of the research revealed fascinating insights into what farmers actually value. A total of 39 farmers participated in the selection process across the two districts, including 19 male-headed and 4 female-headed households in Dawa Cheffa and 16 male-headed households in Ambasel, where no female participants were available. Through extensive group discussions, farmers established their own selection criteria, which included pod number, seed yield, earliness, seed size, seed color, marketability, disease resistance, drought tolerance, uniform maturity, and straw biomass. Using pairwise matrix ranking, farmers then prioritized these criteria. Disease resistance emerged as the single most important trait at both locations, followed by pod number as the second priority. Seed yield and seed color tied for third place, while straw biomass received the lowest ranking, a finding that surprised some researchers who had assumed fodder value would weigh more heavily in farmer decisions.

Direct matrix ranking then allowed farmers to score each variety against each criterion on a scale from one, meaning very good, to five, meaning extremely poor. The results were remarkably consistent across both districts. At Dawa Cheffa, Rasa received the lowest cumulative score of 79, indicating the highest preference, followed by Boreda at 81, Shewa Robit at 105, and NVL-01 at 125. At Ambasel, the pattern repeated almost identically: Rasa scored 65, Boreda 73, NVL-01 114, and Shewa Robit 128. The convergence of scientific yield data and farmer preference rankings on the same two varieties, Rasa and Boreda, represents a powerful validation of both the varieties themselves and the participatory methodology. When farmers and breeders agree, adoption barriers crumble.

Following the selection phase, the research team moved into a pre-scaling-up phase for Rasa, the variety that topped both the yield tables and the farmer preference rankings. Selected farmers received seeds of the chosen variety along with training on recommended seed production and management packages delivered by Sirinka Agricultural Research Center researchers. Development agents and woreda agricultural office experts also participated in the training sessions. A multidisciplinary team of researchers, policymakers, and agricultural experts conducted regular field inspections throughout the growing season, monitoring progress, troubleshooting problems, and advising on everything from land selection through harvest. This phased, hands-on approach to technology dissemination reflects a growing recognition within Ethiopian agricultural research that the gap between variety release and farmer adoption has historically been one of the weakest links in the innovation chain.

The implications of this study extend well beyond the borders of Eastern Amhara. Ethiopia currently produces mung bean on approximately 48,022 hectares nationally, yielding about 51,569 tons, with the Amhara region alone accounting for 68 percent of the national area and 72 percent of national productivity. Yet average yields of 1.073 tons per hectare nationally and 1.133 tons per hectare regionally remain far below the two tons per hectare achievable under experimental conditions. The study demonstrates that closing this gap requires not just better varieties but better processes for getting those varieties into farmers’ hands. By involving farmers directly in evaluation, the researchers accelerated adoption timelines and ensured that the traits farmers actually care about, particularly disease resistance and pod number, were not overlooked in favor of traits that breeders might prioritize in isolation. As climate variability intensifies across the Horn of Africa, the lessons from these Ethiopian mung bean fields offer a template for how participatory approaches can build more resilient and productive agricultural systems from the ground up.

Subject of Research: Participatory varietal selection of mung bean varieties in Eastern Amhara, Ethiopia

Article Title: Participatory selection of mung bean ( Vigna radiata L.) varieties in Eastern Amhara region, Ethiopia

Article References: Tefera, M., Kebede, G., & Haile, A. (2025). Participatory selection of mung bean (Vigna radiata L.) varieties in Eastern Amhara region, Ethiopia. Innovations in Agriculture, 8, 1-8. https://doi.org/10.3897/ia.2025.163481

Image Credits: AI Generated

DOI: 10.3897/ia.2025.163481

Keywords: mung bean, participatory varietal selection, Ethiopia, Amhara region, disease resistance, crop yield, smallholder farmers, Rasa variety, Boreda variety, drought tolerance, pulse crops, agricultural research

PSbMV is a potyvirus that infects a wide range of legume species, including pea, faba bean, lentil, and chickpea. When it is transmitted through seed, the virus can reduce both yield and seed quality, and infected plants serve as inoculum sources for aphid-mediated spread within and between fields. Seed health certification programs rely on laboratory assays such as enzyme-linked immunosorbent assays, reverse transcription polymerase chain reaction, and grow-out tests to detect the pathogen. These methods are accurate but each seed tested is consumed in the process, and screening the large numbers of individual seeds needed to certify a commercial seed lot is expensive and time-consuming. The search for a rapid, non-destructive, and scalable screening technology has therefore been a long-standing goal in seed pathology.

The Saskatchewan research group, led by Simin Sabaghian of the Department of Plant Sciences with collaborators in Mechanical Engineering, hypothesized that viral infection produces measurable changes in the optical properties of seeds. Virus replication alters host metabolism, affecting moisture content, the concentration of phenolic compounds, starch composition, and tissue structure. Each of these biochemical and physical changes can modify how seed tissue reflects and absorbs light across the visible, near-infrared, and shortwave-infrared portions of the electromagnetic spectrum. Hyperspectral imaging captures this information in detail: rather than recording three broad color bands like a conventional camera, a hyperspectral system records a continuous reflectance spectrum for every pixel in an image, allowing researchers to detect subtle spectral fingerprints invisible to the human eye.

In the study, individual faba bean seeds were imaged using two hyperspectral systems covering the visible and near-infrared range together with the shortwave-infrared range, spanning approximately 400 to 1700 nanometers. Directional reflectance spectra were extracted from the calibrated images after preprocessing. The researchers used principal component analysis to explore the data and found that, although infected and healthy seeds are visually indistinguishable, they separate subtly but consistently along principal component axes. This confirmed that infection does leave a detectable optical signature, providing the statistical foundation for building classification models.

The team then trained and compared several supervised machine learning classifiers on the spectral data, testing different preprocessing strategies to maximize discrimination. The best-performing combination was a support vector machine coupled with standard normal variate preprocessing, a technique that normalizes each spectrum to correct for scattering effects and baseline shifts. This model achieved a mean fivefold cross-validation accuracy of 97.2 percent and, critically, an independent holdout accuracy of 98.3 percent, with a receiver operating characteristic area under the curve of 0.994. The near-perfect ROC-AUC indicates that the model separates infected from healthy seeds almost flawlessly across all decision thresholds, a level of performance that suggests the spectral signature of infection is robust rather than an artifact of the training data.

Recognizing that full-spectrum hyperspectral systems are expensive and data-heavy, the researchers also sought to reduce the dimensionality of the problem. Competitive adaptive reweighted sampling, a variable selection algorithm, was used to identify a compact subset of the most informative wavelengths. The selected bands were distributed across the visible, near-infrared, and shortwave-infrared regions, consistent with the idea that multiple biochemical features, including pigmentation, water absorption features, and phenolic-related absorptions, contribute to the classification signal. When the reduced set of wavelengths was used to retrain the model, discrimination remained strong, with cross-validation and holdout accuracies of 96.2 percent and 90.8 percent respectively and a holdout ROC-AUC of 0.974. The modest drop in holdout performance is the expected trade-off of model simplification, but the result demonstrates that a much smaller and cheaper sensor could plausibly deliver near-comparable screening performance in a commercial setting.

To ground the spectral findings in plant biology, the researchers quantified seed moisture content and performed proton nuclear magnetic resonance metabolite profiling on the seeds. The metabolomic analysis revealed elevated phenolic signals in virus-infected seeds, indicating that PSbMV infection triggers a measurable shift in seed secondary metabolism. Phenolic compounds are common components of plant defense responses, and their accumulation likely alters reflectance in both the visible range, where they influence coloration, and the near-infrared and shortwave-infrared ranges, where they modify absorption features. This mechanistic link between infection-induced biochemistry and optical response strengthens the case that the classifier is genuinely detecting disease physiology rather than confounding factors such as seed size, shape, or surface texture.

The practical implications extend across the pulse seed industry. A hyperspectral screening line could potentially evaluate thousands of individual seeds per hour without consuming any of them, allowing certified seed producers to identify and remove infected seeds before planting or sale. Because the seeds remain intact, positive findings could be verified with molecular assays while non-infected seed is preserved. The single-seed resolution is particularly important for seed-borne pathogens, where infection levels are often low and bulked sampling can dilute the signal; a system that scores every seed individually provides a direct estimate of infection incidence within a lot. The Saskatchewan group suggests that the approach could form the foundation for rapid, non-destructive seed health assessment tools and integrated virus management strategies in pulse crops.

Challenges remain before the technology reaches commercial deployment. The models were developed and validated on faba bean seeds under controlled laboratory conditions, and performance will need to be confirmed across different faba bean varieties, growing environments, infection severities, and seed lots to ensure the spectral models generalize beyond the training population. Instrument calibration transfer between hyperspectral cameras, the integration of imaging hardware into high-throughput sorting equipment, and regulatory acceptance of optical screening in certification programs are additional hurdles. Nevertheless, the study provides a rigorous proof of concept, supported by independent validation, wavelength reduction, and metabolomic corroboration, that a virus hiding inside a seed can be exposed by the light it reflects. As pulse production expands globally to meet growing demand for plant protein, tools that protect seed health without sacrificing seed will become increasingly valuable, and this work marks a significant step toward that goal.

Subject of Research: Non-destructive detection of pea seed-borne mosaic virus in faba bean seeds using hyperspectral imaging and machine learning

Article Title: Non-destructive hyperspectral imaging and machine learning for detection of pea seed-borne mosaic virus in faba bean seeds

Article References: Sabaghian, S., Jaliliantabar, F., Noble, S. D., Onu, G., & Prager, S. M. (2026). Non-destructive hyperspectral imaging and machine learning for detection of pea seed-borne mosaic virus in faba bean seeds. Plant Methods. https://doi.org/10.1186/s13007-026-01588-5

Image Credits: AI Generated

DOI: 10.1186/s13007-026-01588-5

Keywords: hyperspectral imaging, machine learning, PSbMV, faba bean, seed health, plant virus detection, support vector machine, non-destructive testing, seed-borne pathogens, pulse crops, NMR metabolomics, spectral classification