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 investigates how cortical microtubule orientations are coordinated across tissues, testing the hypothesis that they sense mechanical stress. By quantifying microtubule behavior on different cell faces and edges, the authors find that tissue-level biases arise from cell‑geometric factors and edge‑specific mechanisms rather than direct stress sensing, and they develop a combinatorial model incorporating face‑specific and edge‑specific microtubule dynamics to explain the observed coordination.

The study isolated the Plant Cysteine Oxidase/Ethylene Response Factor VII oxygen‑sensing circuit from Arabidopsis thaliana and reconstituted it in Saccharomyces cerevisiae, using a reporter to compare hypoxia‑induced transcriptional dynamics in yeast and plants. Both systems showed rapid ERFVII stabilization, but plants exhibited a larger response, which could be enhanced in yeast by adding a hypoxia‑inducible feedback loop. Computational modeling identified promoter competition and hypoxia‑inducible PCOs as key determinants of early hypoxia responses.

The study examined how soil phosphorus and nitrogen availability influence wheat root-associated arbuscular mycorrhizal fungal (AMF) communities and the expression of mycorrhizal nutrient transporters. Field sampling across two years combined with controlled pot experiments showed that P and N jointly affect AMF colonisation, community composition (with Funneliformis dominance under high P), and regulation of phosphate, ammonium, and nitrate transporters. Integrating metabarcoding and RT‑qPCR provides a framework to assess AMF contributions to crop nutrition.

The study investigates the altered timing of the core circadian oscillator gene ELF3 in wheat compared to Arabidopsis, revealing that dawn-specific expression in wheat arises from repression by TOC1. An optimized computational model integrating experimental expression data and promoter architecture predicts that wheat’s circadian oscillator remains robust despite this shift, indicating flexibility in plant circadian network design.

The study investigated unexpected leaf spot symptoms in Psa3‑resistant kiwifruit (Actinidia) germplasm, finding that Psa3 was detectable by qPCR and metabarcoding despite poor culturing. Metabarcoding revealed distinct bacterial community shifts in lesions versus healthy tissue, and whole‑genome sequencing identified diverse Pseudomonas spp. that, while not individually more pathogenic, could enhance Psa3 growth, suggesting pathogenic consortia on resistant hosts.