The study introduces a hybrid modeling framework that integrates a logistic ordinary differential equation with a Long Short-Term Memory neural network to form a Physics-Informed Neural Network (PINN) for predicting wheat plant height. Using only time and temperature as inputs, the PINN outperformed other longitudinal growth models, achieving the lowest average RMSE and reduced variability across multiple random initializations. The results suggest that embedding biological growth constraints within data‑driven models can substantially improve prediction accuracy for plant traits.

The study examines how ectopic accumulation of methionine in Arabidopsis thaliana leaves, driven by a deregulated AtCGS transgene under a seed‑specific promoter, reshapes metabolism, gene expression, and DNA methylation. High‑methionine lines exhibit increased amino acids and sugars, activation of stress‑hormone pathways, and reduced expression of DNA methyltransferases, while low‑methionine lines show heightened non‑CG methylation without major transcriptional changes. Integrated transcriptomic and methylomic analyses reveal a feedback loop linking sulfur‑carbon metabolism, stress adaptation, and epigenetic regulation.

The study shows that the SnRK1 catalytic subunit KIN10 directs tissue-specific growth‑defense programs in Arabidopsis thaliana by reshaping transcriptomes. kin10 knockout mutants exhibit altered root transcription, reduced root growth, and weakened defense against Pseudomonas syringae, whereas KIN10 overexpression activates shoot defense pathways, increasing ROS and salicylic acid signaling at the cost of growth.

The study examined how genetic variation among 181 wheat (Triticum aestivum) lines influences root endophytic fungal communities using ITS2 metabarcoding. Heritability estimates and GWAS identified 11 QTLs linked to fungal clade composition, highlighting genetic control of mycobiota, especially for biotrophic AMF. These findings suggest breeding can be used to modulate beneficial root-fungal associations.