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.

The review examines the genetic networks governing spikelet number per spike (SNS) in wheat, highlighting how the balance between inflorescence meristem activity and the timing of terminal spikelet transition determines yield potential. It discusses how mutations affecting meristem identity can create supernumerary spikelets, the trade-offs of such traits, and recent advances using spatial transcriptomics, single‑cell analyses, and multi‑omics to identify new SNS genes for breeding.

The study reveals that each individual plant possesses a statistically unique leaf appearance that can be discriminated using convolutional neural network (CNN) based deep learning, enabling a "plant face" recognition concept. Applications demonstrated include distinguishing leaves from the same species/cultivar, analyzing leaflet positional patterns on compound leaves, assessing bilateral symmetry, and detecting morphological differences linked to stem chirality, highlighting the encoding of genetic, environmental, and developmental information in leaf morphology.

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 applied dual-species spatial transcriptomics at single-cell resolution to map plant and fungal gene activity in rice roots colonized by Rhizophagus irregularis, revealing transcriptional heterogeneity among morphologically similar arbuscules. By pioneering an AM-inducible TRAP-seq using stage‑specific promoters, the authors uncovered stage‑specific reprogramming of nutrient transporters and defence genes, indicating dynamic regulation of nutrient exchange and arbuscule lifecycle.

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.

The study examined how polyploidy, hybridization, and incomplete lineage sorting affect phylogenetic reconstructions in the genus Packera, evaluating several published paralog‑processing pipelines. Results showed that the choice of orthology and paralog handling methods markedly altered tree topology, time‑calibrated phylogenies, biogeographic histories, and detection of ancient reticulation, underscoring the need for careful methodological selection alongside comprehensive taxon sampling.

The study evaluated whether integrating genomic, transcriptomic, and drone-derived phenomic data improves prediction of 129 maize traits across nine environments, using both linear (rrBLUP) and nonlinear (SVR) models. Multi-omics models consistently outperformed single-omics models, with transcriptomic data especially enhancing cross‑environment predictions and capturing genotype‑by‑environment interactions. The results highlight the added value of combining transcriptomics and phenomics with genotypes for more accurate and generalizable trait prediction in maize.

The study introduces the Botanical Spectrum Analyzer (BSA), a GUI that incorporates a modified U‑Net deep neural network for accurate segmentation of plant images from RGB and hyperspectral (VNIR and SWIR) data. BSA was tested on wheat, barley, and Arabidopsis datasets, achieving >99% accuracy and F1‑scores above 98%, and markedly outperformed commercial tools on root segmentation tasks.