The study used transcriptomic profiling to compare tomato (Solanum lycopersicum) responses to three evolutionarily distant pathogens—nematodes, aphids, and oomycetes—during compatible interactions, identifying differentially expressed genes and key host hubs. Integrating public datasets and performing co‑expression and GO enrichment analyses, the authors mapped shared dysregulation clusters and employed Arabidopsis interactome data to place tomato candidates within broader networks, highlighting potential targets for multi‑pathogen resistance.

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 identifies an autophagy pathway that degrades plasma membrane-derived clathrin-coated vesicles (CCVs) during hyperosmotic stress, helping maintain membrane tension as cell volume decreases. Using live imaging and correlative microscopy, the authors show that the TPLATE complex subunits AtEH1/Pan1 and AtEH2/Pan1 act as selective autophagy receptors by directly binding ATG8, thereby removing excess membrane under drought or salt conditions.