Phosphite (Phi) and phosphate (Pi) share the same root uptake system, but Phi acts as a biostimulant that modulates plant growth and disease resistance in a species‑ and Pi‑dependent manner. In Arabidopsis, Phi induces hypersensitive‑like cell death and enhances resistance to Plectosphaerella cucumerina, while in rice it counteracts Pi‑induced susceptibility to Magnaporthe oryzae and Fusarium fujikuroi, accompanied by extensive transcriptional reprogramming.

The authors applied semi‑supervised deep‑learning to super‑resolution images of modern and fossil grass pollen, training convolutional neural networks to extract abstract morphological features. These features were used to quantify past grass community diversity and C3:C4 ratios in a 25,000‑year lake‑sediment record, revealing a marked diversity loss during the last glacial and a gradual decline of C4 grasses in the Holocene.

The authors introduce AdaPoinTr, a geometry-aware transformer that predicts the alpha‑shape of coniferous tree crowns from incomplete terrestrial or mobile laser‑scanning point clouds, focusing on crown reconstruction rather than full tree completion. Trained on synthetically generated partial crowns, the model consistently improves crown shape similarity and reduces height estimation bias across three diverse forest datasets, providing a cost‑effective solution for enhanced 3D forest structural monitoring.

Using diverse Mexican maize varieties and a MAGIC population, the study demonstrated that leaf vein density is both variable and plastic, correlating positively with photosynthetic rates for small intermediate veins and increasing under heat in drought-adapted lines. Twelve QTLs linked to vein patterning were identified, highlighting candidate genes for intermediate vein development and shedding light on the evolution of high-efficiency C4 leaf architecture.

The study assessed 17 morphological, biochemical, and salt‑stress tolerance traits in 19 maize (Zea mays) landrace accessions from northern Argentina, revealing substantial variation both within and among accessions. Redundancy analysis linked phenotypic variation to the altitude of the collection sites, underscoring the potential of these landraces as sources of diverse biochemical and stress‑related traits for breeding.

The study tests whether heavy‑metal stress contributed to maize domestication by exposing teosinte (Zea mays ssp. parviglumis) and the Palomero toluqueno landrace to sublethal copper and cadmium, then analysing genetic diversity, selection signatures, and transcriptomic responses of three chromosome‑5 heavy‑metal response genes (ZmHMA1, ZmHMA7, ZmSKUs5). Results reveal strong positive selection on these genes, heavy‑metal‑induced phenotypes resembling modern maize, and up‑regulation of Tb1, supporting a role for volcanic‑derived metal stress in early maize evolution.

The study examined how DNA methylation influences cold stress priming in Arabidopsis thaliana, revealing that primed plants exhibit distinct gene expression and methylation patterns compared to non-primed plants. DNA methylation mutants, especially met1 lacking CG methylation, showed altered cold memory and misregulation of the CBF gene cluster, indicating that methylation ensures transcriptional precision during stress recall.

The study uncovers how arbuscular mycorrhizal (AM) fungi induce lateral root formation in maize by activating ethylene‑responsive transcription factors (ERFs) that regulate pericycle cell division and reshape flavonoid metabolism, lowering inhibitory flavonols. It also shows that the rhizobacterium Massilia collaborates with AM fungi, degrading flavonoids and supplying auxin, thereby creating an integrated ethylene‑flavonoid‑microbe signaling network that can be harnessed to improve nutrient uptake and crop sustainability.

The study assessed how well common deep learning models (ResNet, EfficientNet, Inception, MobileNet) generalize across different tomato pest and disease image datasets. While models performed well on the dataset they were trained on, they suffered substantial accuracy drops when applied to other datasets, indicating that architectural changes alone cannot overcome dataset variability. The results highlight the necessity for more diverse, representative training data to improve real-world deployment of PPD diagnostic tools.

The study demonstrates that hyperspectral imaging can non‑destructively differentiate active nitrogen‑fixing root nodules from non‑fixing nodules and root tissue based on distinct spectral signatures. By integrating deep‑learning models, the authors created an automated nodule counting pipeline that works across multiple legume species and growth conditions, eliminating labor‑intensive manual counting and reliably detecting nodules within dense root systems.