Phenotypic scoring of Canola Blackleg severity using machine learning image analysis
Authors: Hu, Q., Anderson, S. N., Gardner, S., Ernst, T. W., Koscielny, C. B., Bahia, N. S., Johnson, C. G., Jarvis, A. C., Hynek, J., Coles, N., Falak, I., Charne, D. R., Ruidiaz, M. E., Linares, J. N., Mazis, A., Stanton, D. J.
The study introduces a deep‑learning based image analysis pipeline that scores blackleg disease severity from stem cross‑section images of canola species, achieving greater consistency than median expert raters while preserving comparable heritability of susceptibility traits. This standardized scoring method aims to improve selection of resistant varieties in breeding programs.
The study presents a deep‑learning pipeline that uses state‑of‑the‑art convolutional neural networks to automatically estimate the establishment of perennial groundcovers in agricultural research plots from smartphone images. By employing region‑of‑interest markers and deploying the models on AWS SageMaker with a lightweight Django web interface, the approach provides fast, objective, and reproducible assessments that can be adopted by researchers and growers across the Midwest.
The study generated a phenotypic dataset for 550 Lactuca accessions, including 20 wild relatives, and applied an iterative two‑step GWAS using a jointly processed SNP set for cultivated lettuce (L. sativa) and its wild progenitor (L. serriola) to dissect trait loci. Known and novel QTLs for anthocyanin accumulation, leaf morphology, and pathogen resistance were identified, with several L. serriola‑specific QTLs revealing unique genetic architectures, underscoring the breeding value of wild lettuce species.
The study used chlorophyll fluorescence imaging to map non-photochemical quenching (NPQ) gradients along barley leaf axes and found heat stress attenuates NPQ induction, revealing spatial heterogeneity in stress responses. Genome‑wide association and transcriptomic analyses identified candidate genes, notably HORVU.MOREX.r3.3HG0262630, that mediate region‑specific heat responses, highlighting pathways for improving cereal heat resilience.
Unraveling the cis-regulatory code controlling abscisic acid-dependent gene expression in Arabidopsis using deep learning
Authors: Opdebeeck, H., Smet, D., Thierens, S., Minne, M., De Beukelaer, H., Zuallaert, J., Van Bel, M., Van Montagu, M., Degroeve, S., De Rybel, B., Vandepoele, K.
The study used an interpretable convolutional neural network to predict ABA responsiveness from proximal promoter sequences in Arabidopsis thaliana, revealing both known ABF-binding motifs and novel regulatory elements. Model performance was boosted by advanced data augmentation, and predicted regulatory regions were experimentally validated using reporter lines, confirming the inferred cis‑regulatory code.
The study applied Spatial Analysis of Field Trials with Splines (SpATS) and Neighbor Genome-Wide Association Study (Neighbor GWAS) to barley field data, revealing that neighboring genotypes contribute to spatial variation in disease damage. Neighbor GWAS identified variants on chromosome 7H that modestly affect net form net blotch and scald resistance, suggesting that genotype mixtures could mitigate pest damage.
The study presents GenoRetriever, an interpretable deep learning framework trained on STRIPE-seq data from soybean and other crops, that predicts transcription start site locations and usage by identifying 27 core promoter motifs. Validation using in silico motif insertions, saturation mutagenesis, and CRISPR‑Cas9 promoter editing demonstrates high predictive accuracy and reveals domestication‑driven motif usage shifts and lineage‑specific effects. The tool is provided via a web server for promoter analysis and design, offering a new resource for plant functional genomics and crop improvement.
The study evaluated natural genetic variation in non-photochemical quenching and photoprotection across 861 sorghum accessions grown in the field over two years, revealing moderate to high broad-sense heritability for chlorophyll fluorescence traits. By integrating genome-wide association studies (GWAS) with transcriptome-wide association studies (TWAS) and covariance analyses, the authors identified 110 high-confidence candidate genes underlying photoprotection, highlighting a complex, polygenic architecture for these traits.
The study used comparative transcriptomics across Erysimum species to identify two 2‑oxoglutarate‑dependent dioxygenases, CARD5 and CARD6, responsible for the 14β‑ and 21‑hydroxylation steps in cardenolide biosynthesis in Erysimum cheiranthoides. Knockout mutants lacking these genes accumulated pathway intermediates, and transient expression in Nicotiana benthamiana confirmed their enzymatic functions, while structural modeling pinpointed residues linked to neofunctionalization.
Ethylene and ROS Signaling Are Key Regulators of Lateral Root Development under Salt Stress in Tomato
Authors: Rahmati Ishka, M., Zhao, J., Sussman, H., Mohanty, D., Craft, E., Yu, L., Pineros, M., Tester, M., Kawa, D., Mittler, R., Nelson, A., Fei, Z., Julkowska, M. M.
The study examined salt-induced alterations in root system architecture across a diverse panel of wild and cultivated tomato accessions, identifying tolerant varieties with distinct lateral root strategies. By combining Bulk Segregant Analysis of an F2 population with GWAS, the authors pinpointed 22 candidate genes, further narrowing to two key regulators through RNA‑Seq and functional assays involving ethylene and ROS profiling. These findings reveal genetic targets for improving salt resilience in tomato root development.