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 used a computer‑vision phenotyping pipeline (EarVision.v2) based on Faster R-CNN to map Ds‑GFP mutant kernels on maize ears and a statistical framework (EarScape) to assess spatial patterns of allele transmission from the apex to the base. They found that alleles causing pollen‑specific transmission defects often show significant spatial biases, whereas Mendelian alleles do not, indicating that reduced pollen fitness can shape the spatial distribution of progeny genotypes in Zea mays.