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.

Two short‑cycle dwarf eggplant lines, Micro‑Mel and Mini‑Mel, were developed through introgressions from Solanum anguivi and characterized for growth, phenology, and compact architecture, enabling up to three seed‑to‑seed cycles per year. In vitro regeneration assays and 23× whole‑genome sequencing revealed low heterozygosity, introgressed genome segments, and dwarfism‑related orthologs, establishing these lines as rapid, space‑efficient model systems for eggplant genetics and breeding.

RNA‑seq of 328 wheat lines using a pan‑genome reference uncovered over 20,000 additional transcripts beyond the Chinese Spring genome and enabled construction of a pan‑gene eQTL regulatory atlas. Multi‑omics integration identified 231 high‑confidence candidate genes influencing 34 agronomic traits and powdery mildew resistance, with functional validation showing 80% of candidates affecting trait phenotypes via an EMS mutant library.