New Phytologist

Quantifying root traits such as root length (RL) and root surface area (RSA) from minirhizotron imagery is a valuable approach for overcoming the phenotyping bottleneck that limits understanding and improvement of crop productivity, resource use efficiency and resilience in field experiments. However, current approaches remain labor-intensive, and deep learning (DL) methods suffer from limited generalization ability. We present RootQuant, an end-to-end DL model that simultaneously predicts RL and RSA directly from minirhizotron images using only whole-image trait values as supervision, thereby eliminating the need for pixel-level annotations. The models generalization ability was evaluated across species and fine-tuning configurations. The practical applicability of the model was further assessed under field conditions by converting image-derived RL estimates into volumetric root length density (vRLD). Using 118,191 maize and soybean images collected between 2009 and 2020, RootQuant trained on both species achieved an R2 of 0.90 and an RMSE of 2.9 mm for RL, and an R2 of 0.88 and an RMSE of 4.2 mm2 for RSA. The same mixed-species model generalized strongly across species, yielding an 8% relative improvement in R2 and a 30% lower RMSE on maize compared with the same architecture trained on a single species and applied zero-shot. Image-derived RL predictions converted to vRLD showed the expected depth-dependent decline in vRLD, as was also found by coincident destructive quantification of roots washed out of soil cores. By providing a generalist backbone model trained on a large dataset from two major crop species, RootQuant enables high-throughput simultaneous estimation of two relevant root traits directly from raw imagery without task-specific fine-tuning, thereby accelerating in situ root system analysis and phenotyping applications.

Adaptation of plant photosynthesis to dynamic light conditions experienced in natural environments is achieved through specific protective mechanisms. Energy-dependent non-photochemical quenching (qE), regulated by Photosystem II Subunit S (PsbS), is a key process facilitating acclimation to fluctuating light in C3 plants, which operate conventional photosynthesis. C4 plants, which include some of the world's most productive and agriculturally important crops, have evolved a distinct high-efficiency photosynthetic pathway. Little is known about the role of specific processes, like qE, in acclimation of C4 plants to dynamic light environments. We generated gene-edited lines of the model C4 grass Setaria viridis lacking PsbS, which were found to be deficient in qE. This deficiency resulted in a modest increase in PSII photoinhibition and a CO2 assimilation penalty under light stress in short-term experiments, but photosynthesis and growth under fluctuating light were unaffected. Instead, keeping Photosystem I oxidised through photosynthetic control, negative feedback regulation of the Cytochrome b6f complex, was critical. Therefore, unlike in C3 plants, qE does not provide a significant adaptive advantage to C4 plants under dynamic light conditions. These findings provide important insights into the biology of C4 plants and help prioritise future strategies for improving the productivity and resilience of C4 crops.

Rafflesiaceae, known for producing the largest flowers in the world, are obligate parasites that exclusively infect Tetrastigma sp. (Vitaceae). Despite their unique biology, the interactions between parasitic tissues and host roots remain poorly understood, particularly during the flower morphogenesis phase, where parasitic tissue erupts through the host root. Here, we performed dual transcriptome analyses of two Rafflesiaceae species and their respective Tetrastigma hosts: Sapria himalayana with T. cauliflorum and Rafflesia speciosa with T. magnum. Our findings reveal species-specific transcriptional responses in Tetrastigma, suggesting divergent parasitism strategies between Rafflesia and Sapria. Moreover, we identify molecular signatures of parasitism that parallel plant gall formation, particularly in genes governing cell wall modification and host tissue reorganization. Unlike bacterial or insect-induced galls, these mechanisms may involve fungal symbionts, highlighting the unique nature of these interactions. Together, our results demonstrate that Rafflesiaceae parasitism represents a complex tripartite relationship among host, holoparasite, and associated microbes, offering new insights into the hidden biology of these remarkable parasitic plants.

Phenotyping stomatal traits and their developmental plasticity is time-consuming but holds potential to improve water use efficiency and photosynthesis for designing stress-tolerant crops under climate change. Here, we develop a robust, high-throughput pipeline for phenotyping 14 stomatal traits in winter wheat related to size, variation, maximum conductance, and spatial patterning. We (1) analyze over 25,000 images from 60 wheat cultivars grown in growth chamber, greenhouse, and field conditions; (2) investigate the impact of light, temperature, and reduced water and nitrogen supply on stomatal traits and their developmental plasticity across adaxial and abaxial surfaces; and (3) evaluate genetic diversity and breeding progress of stomatal traits. Stomatal traits were highly broad-sense heritable, were largely plastic in response to environmental conditions, and showed genotype-specific responses. Stomatal traits of third leaves under controlled environments with stable light and temperature conditions reliably captured the genetic variance of flag leaves under field conditions. Our data suggests that the upper leaf surface contributed more to transpiration and cooling through consistently higher stomatal density, area, and maximum conductance, while the lower surface facilitated CO2 diffusion via systematic proper patterning and spacing. Breeding maintains the genetic diversity of stomatal traits, and our pipeline facilitates breeders to target them to enhance water use efficiency in high-yielding modern cultivars.

Biological invasions are transforming ecosystems worldwide, yet the genomic bases enabling certain species to dominate new environments remain poorly understood. Phragmites australis, a widespread wetland grass with invasive and native subspecies co-occurring in North America, provides a powerful system to investigate genomic mechanisms of invasiveness. We generated independent chromosome-scale genome assemblies for invasive P. australis ssp. australis and co-occurring native ssp. americanus and used comparative genomic and transcriptomic analyses to identify lineage-specific innovations associated with invasive success. The invasive subspecies exhibits genomic novelties through functionally-biased single-copy orthologs, intronless genes, and subgenome expression asymmetry, along with a stress-ready basal transcriptome relative to the native subspecies. Following the removal of aboveground shoots ("cutback"), which measures the ability to recover from damage, the invasive subspecies undergoes stronger transcriptional reprogramming, increased shoot production, and higher biomass accumulation compared to the native. It also displays expansion of gene families and coordinately expressed gene modules that support resource mobilization, growth responses to light, and stress tolerance. Beyond Phragmites, comparative analyses across multiple grass genomes, including eight invasive species with related non-invasive species, revealed repeated expansion of gene families associated with abiotic stress tolerance and developmental regulation, suggesting convergent adaptive strategies in the grass family for invasive success. Together, these results demonstrate genomic architecture linked to invasion success and highlight potential targets for managing invasive grasses.

Accurate segmentation of plant disease symptoms is essential for crop monitoring and phenotyping, yet it typically requires costly pixel-level annotations. Weakly supervised semantic segmentation (WSSS) alleviates this burden using image-level labels, but its performance depends on the quality of spatial priors such as class activation maps (CAMs). We investigate whether text-guided segmentation with the Segment Anything Model 3 (SAM3) can serve as an alternative weak supervision signal. Three pseudo-mask generation strategies are compared: (i) CAMs refined with SAM or SAM3, (ii) zero-shot text-guided SAM3, and (iii) a hybrid approach combining weak spatial cues with text prompts. The resulting pseudo-masks are used to train a DeepLabV3 model. Text guidance alone matches or outperforms conventional WSSS, achieving up to 0.46 IoU without spatial supervision and 0.61 IoU on a public dataset, although performance is sensitive to text prompt formulation. The hybrid strategy improves robustness, reaching 0.50 IoU on the primary dataset and 0.58 IoU on the additional dataset while reducing prompt sensitivity. Overall, text guidance is a promising alternative to conventional weak supervision, while hybrid approaches provide a more robust solution for plant disease segmentation.

While whole genome sequencing captures millions of single nucleotide polymorphisms (SNPs) and hyperspectral imaging (HSI) enables non destructive plant phenotyping, integrating these modalities to link genotype to phenotype remains challenging due to their high dimensionality and non linearity. This study presents DeepPheno a deep learning framework that predicts SNP genotypes from HSI data, using model predictability as a proxy for genotype phenotype association. HSI data were acquired from 194 lettuce genotypes under field conditions. HSI data patches (20 x 20 pixels x 224 spectral bands) were used to train a hybrid CNN to predict the variant of a specific SNP. The framework was validated on SNPs with known phenotypic effects (anthocyanin, leaf serration, pale pigmentation), achieving high predictive performance (AUC ranging from 0.806 to 0.935), whereas models trained on randomly shuffled labels performed at chance (mean AUC {approx} 0.51). Extending the workflow to 50 randomly selected putatively neutral SNPs, most yielded low predictability, but two showed high performance (AUC > 0.76), suggesting uncharacterized genotype phenotype links. Explainable AI, including SHAP and Grad CAM, identified relevant spectral and spatial features driving these predictions, particularly the green and red edge wavelengths associated with pigment dynamics and leaf structure. These results establish a framework for understanding complex genotype phenotype interactions in plants and extracting these links from HSI data without predefining the exact trait values. It provides an avenue for high throughput trait discovery and description and extends the integration of image based phenomics with plant genetics.

Shifts in flowering phenology are one of the most well studied plant responses to global climate change. Many studies have documented these shifts and their drivers, including some that describe altered patterns of co-flowering among taxonomically broad species within communities. In comparison, few analyses have examined systematic changes in co-flowering between closely related, interfertile species, where co-flowering can have unique evolutionary consequences. To address such shifts in co-flowering among close relatives, we investigate phenological responses to climate change and its effect on patterns of co-flowering over the past 124 years in 52 species of North America violets (Viola). This genus has many co-occurring species that reproductively interact via shared pollinators and hybridization. We use ~14,000 herbarium records along with environmental and species trait data to model the magnitude of recent flowering phenology shifts, environmental variables and/or species traits associated with these shifts, and resulting changes in co-flowering among species. While both the magnitude and direction of phenological shifts varied among Viola species, nearly half (25/52) show significant changes in flowering day. Regardless of whether flowering was advanced or delayed, flowering date was most consistently associated with local mean temperature. Of six species-level traits, geographical region also significantly predicted flowering shifts, consistent with environment and geography together explaining broad phenological responses across this group. These shifts have produced significant changes to pairwise patterns of co-flowering among species -- ranging from a 59 day increase in co-flowering to complete loss of co-flowering overlap. Sympatric pairs specifically have experienced both increases and decreases in co-flowering, with a geographic pattern of increased co-flowering occurring mainly in eastern US and decreased co-flowering common in western US. Because Viola species are generalist pollinated and already known to hybridize, these new co-flowering patterns could further undermine reproductive barriers among species in this genus.

O_LICereal grafting using embryonic tissues has recently become technically feasible; however, the physiological consequences of cereal grafting remain uncharacterized. C_LIO_LIWe systematically evaluate photosynthetic performance and stomatal dynamics across different graft combinations in two photosynthetically distinct species, rice (C3) and pearl millet (C4). We first assessed steady-state photosynthetic performance and dynamic stomatal responses in five-week-old rice and pearl millet grafts grown under a saturated water regime, to establish whether cereal grafting alters physiology at early stages. Next, we assessed same traits at the onset of optimal water regime, and after five days to determine whether any graft-induced effects on photosynthesis or growth persisted over time. C_LIO_LIWe observed that across contrasting water regimes and at different plant developmental stages, cereal grafting did not alter growth, photosynthesis or stomatal kinetics in either species, while revealing modest early stage C4-specific adjustments in stomatal dynamics without affecting photosynthetic capacity or biochemical parameters. C_LIO_LIWe demonstrate that cereal grafting does not alter core physiological traits in rice or pearl millet and can be deployed without long-term impact on photosynthesis. These findings establish cereal grafting as a tractable platform for mechanistic dissection of root-shoot signaling and trait combination across different C3 and C4 cereals. C_LI

In Asteraceae species, the capitulum is a compact inflorescence, featuring a characteristic reproductive structure. Despite the identification of a few key regulatory factors, the transcriptome-level information on the developing capitulum remains limited. Here, we applied single-cell and spatial transcriptome sequencing to investigate the developing Gerbera hybridas capitulum during floret differentiation. We obtained a transcriptomics atlas encompassing different stages of the Gerbera capitulum and analyzed the cellular and spatial dynamics of gene expression. Using marker gene expression and GO enrichment of cluster-specific DEGs, we annotated putative cell types and described changes in gene expression across sampled stages, potentially associated with ongoing developmental processes. We detected activity of previously undescribed MADS-box genes and defined their spatial expression patterns. Notably, the MADS-box gene GAGL12 was found to be enriched in the putative capitulum phloem cells. The GAGL12 protein was shown in yeast two-hybrid assays to interact with several other MADS-domain proteins with hypothesized functions in vasculature development, and further detailed in silico analyses supported a candidate role in the development of capitulum vasculature. Altogether, we provide integrative and dynamic transcriptomic insight into capitulum and floret development and lay a basis for future functional studies of the control and development of this intriguing reproductive structure.

O_LIPhytoplankton species display characteristic morphologies that are generally assumed to confer adaptive advantages, yet the functional significance of cell shape remains poorly understood. Here, we investigated whether pleomorphism contributes to acclimation to changing light environments. C_LIO_LIUsing the cyanobacterium Anabaena sp. PCC 7120 as a model system, we combined molecular genetics, microscopy, physiological measurements and biophysical analyses to determine how morphology is regulated and how it affects photosynthetic performance under different light intensities. C_LIO_LIWe show that Anabaena undergoes a reversible light-dependent morphological transition from rod-shaped cells under low light to large globular cells under high light stress. This transition is controlled by the relative activities of the elongasome and class A penicillin-binding proteins and is accompanied by thylakoid reorganization. The globular morphology reduces light absorption and enables cells to maintain photosynthetic activity under photoinhibitory conditions. C_LIO_LIOur findings establish a mechanistic link between cell-wall remodelling, cellular optics and photosynthetic performance, revealing pleomorphism as a dynamic acclimation strategy to high light stress. More broadly, this work provides experimental support for the packaging effect and highlights morphology as an active determinant of phytoplankton fitness. C_LI

Flowering time strongly influences crop adaptation, plant architecture, generation turnover, and breeding efficiency, but the functional organization of florigen genes remains poorly resolved in many polyploid orphan crops. Quinoa (Chenopodium quinoa) is a climate-resilient allotetraploid crop with extensive variation in flowering behavior, and genome analyses have identified multiple FLOWERING LOCUS T (FT)-like homologs. However, genome sequence and expression information alone cannot determine which homologs provide effective florigenic output in planta. Here, we combined apple latent spherical virus-mediated overexpression (VOX) and virus-induced gene silencing (VIGS) in quinoa with heterologous expression in Arabidopsis thaliana, domain-swapping analyses, and cross-germplasm validation to functionally dissect quinoa FT activity. Although several CqFT homologs were transcriptionally induced during the floral transition, their functional outputs were markedly unequal. CqFT1A and CqFT1B-1 acted as the major florigenic activators: overexpression of either gene induced rapid and synchronized flowering, whereas CqFT1-VIGS delayed flowering. In contrast, CqFT2A and CqFT2B retained only weak flowering-promoting activity, whereas CqFT1B-2 showed no detectable promotive effect under the conditions tested, revealing a clear functional hierarchy among transcriptionally induced CqFT homologs. Domain-swapping analyses showed that C-terminal variation contributes to, but does not fully explain, functional divergence among CqFT homologs. In late-flowering highland lines, elevated FT input accelerated flowering, induced coordinated floral transition, and shortened the time to viable seed production. These findings identify CqFT1A and CqFT1B-1 as the major florigenic activators in quinoa and establish a functional genomics framework for resolving and modulating flowering-time control in polyploid orphan crops.

The emergence of vascular plants on land is one of evolution greatest triumphs. This success was contingent on the capacity of roots and their symbionts to acquire resources from exposed geology. However, how rock chemistry shapes plant root architectural strategies, and their return on investment during early ecosystem colonization remains poorly understood. Here we use a two-year mesocosm experiment with Bouteloua dactyloides grass and an arbuscular mycorrhizal symbiont, grown on four mineral substrates of contrasting composition, to show that rock geochemistry predictably determines root topological strategy, from herringbone architecture on nutrient-poor granite to dichotomous-like branching on nutrient-rich basalt. Substrate identity governed investment allocation between root complexity and biomass, with plants consolidating existing transport pathways as weathering-derived nutrients subsided. Traits associated with exploratory effort were generally decoupled from those related to biomass buildup. In basalt and rhyolite plants preferentially invested in complexity, generating the largest numbers of prospective tips for mining and biomass buildup; in granite, plants chose a surviving strategy, limiting branching to preserve biomass; while in schist, plants balanced biomass with complexity, extending growth on low investment, which increased tissue density. Surprisingly, mycorrhizal fungi did not alter the whole root system size, but reallocated investment between specific root orders, discouraging investment in embryonic roots in some substrates, and stimulating lateral expansion of the rooting system in others. This extends the functional balance mechanism from plant to the plant-fungus system. The extensive phenotypic plasticity of the root-mycorrhiza system shown here provides an evolutionary space for natural selection, which must have played a crucial role in the success of plants on land in the past, and is crucial for understanding plant ecological dynamics today.

The climatic transition from Marine Isotope Stage 3 (MIS3) to the Last Glacial Maximum (LGM) had caused widespread vegetation change. Despite the dynamic equilibrium between vegetation and climate, the specific role of functional composition in vegetation response to climate change was inadequately understood. Here, we analyzed the long-term trajectories of palynological diversity, vegetation coverage and community-weighted-mean (CWM) functional traits based on EH22 pollen record (35-18 cal ka BP) from Erhai Lake, southwestern China. The results disclosed a vegetation transition from temperate deciduous broadleaf forest dominance in late MIS3 to cold coniferous and mixed broadleaved/coniferous forests in LGM. This vegetation dynamic involved functional composition shifts from competitive-driven functional convergence to partial recovery via niche differentiation during the late MIS3, and finally to a low-diversity but functional differentiation state through trait complementarity and diversification strategies during the cold LGM. Our results likely support a function-mediated climate filtering process whereby climate change regulated long-term vegetation dynamics during the MIS3 to LGM transition primarily through shifts in CWM functional composition. These findings underscore the potential of pollen-based trait approaches to reconstruct ecosystem properties and advance our understanding of ecosystem change over decadal to millennial time-scales.

Grasslands are vulnerable to increasing drought with global warming, but process-based models lack the mechanistic knowledge required to predict the magnitude of drought impacts. While a plant hydraulics framework has been successful in advancing process understanding of drought responses in trees, and how drought responses vary across rainfall gradients, similar approaches have rarely been applied to grasses. Here, we quantified the progression of key drought response processes in sixteen dominant perennial grasses (seven C3 and nine C4) with differing climatic origins across eastern Australia. We found that stomatal closure, hydraulic impairment and leaf browning occurred concurrently, in contrast to the progressive sequence typically observed in trees. We also found that drought response traits were not correlated with species climate of origin. The early impairment of leaf hydraulic conductance and leaf browning along with the lack of correlation with climate of origin suggest that grasses may employ fundamentally different strategies to adapt to low water availability than trees. These results highlight the need for grass-specific parameterization of drought responses in process-based models.

Seed quality is a complex trait shaped by morphological, biochemical and microbiological properties that are rarely characterised simultaneously, limiting our ability to identify robust predictive indicators of germination speed and seedling emergence across varieties. Here, we performed a multi-factor characterisation of eight common bean (Phaseolus vulgaris L.) varieties, combining seed morphometrics, untargeted GC-MS metabolomics on three seed organs, and amplicon sequencing of bacterial and fungal communities, to identify indicators of germination speed and emergence percentage. The eight varieties showed substantial variation in both traits, used as physiological seed quality proxies. Seed weight and size variation between varieties were correlated with germination speed. The intravariety variance of seed weight was independently correlated with emergence performance. Metabolome composition differed strongly across seed organs, with variety as the dominant driver. Individual-seed metabolomic profiles in the plumule and cotyledon were associated with germination speed but not emergence, yielding 16 plumule and three cotyledon candidate metabolite markers. Fungal community composition was associated with both germination speed and emergence, while bacterial communities were associated with emergence only. Nine fungal and four bacterial taxa were identified as candidate indicators. Inter-kingdom co-occurrence network analysis revealed that fungi with similar germination speed associations tend to cluster in the same modules, suggesting that community-level modules rather than individual taxa may constitute more robust microbial indicators. These results demonstrate that germination speed and emergence capacity are governed by distinct seed properties, and provide morphological, metabolic and microbial candidate indicators for integration into targeted seed quality assessment frameworks for common bean.

Green algae must continuously balance resource availability to maintain photosynthetic performance. The O2:CO2 ratio is a key determinant of their metabolic mode. Under hyperoxia or low CO2, many algae induce a carbon concentrating mechanism (CCM). In the model green alga Chlamydomonas reinhardtii, the CCM relies on a pyrenoid, a specialized microcompartment that elevates CO2 around rubisco. While ambient CO2 acclimation is well-studied, responses to hyperoxia remain poorly understood, despite its frequent occurrence in nature under high light. Using controlled bioreactors, we exposed two diverse Chlamydomonas ecotypes, CC1009 and CC2343, to 95% oxygen to analyze time-dependent, genome-wide transcriptomic and phenotypic changes. Both ecotypes induced CCM genes, but they exhibited distinct molecular and physiological phenotypes. The tolerant ecotype (CC1009) successfully adapted, developing a functional CCM with a structured starch sheath. Conversely, the sensitive ecotype (CC2343) suffered growth arrest and formed malformed pyrenoids. Transcriptomics revealed that CC1009 initiated a rapid initial response, upregulating chloroplast proteostasis and downregulating nucleotide metabolism. CC2343 showed a massive, delayed transcriptional response, downregulating genes coding for photosystems and tetrapyrrole biosynthesis. This unbiased transcriptomic approach identifies key candidate genes driving algal acclimation to hyperoxic stress in natural, high-light environments.

Incomplete information on distributions for a high proportion of the world's plant species together with biases in global biodiversity data mean that current estimates of plant diversity patterns are skewed. A key issue is that current predictions rely on a subset of species that is not representative of all plant species. Here we tested the feasibility of a representative sampling approach for mapping global vascular plant diversity at the finest scale where comprehensive data is available. Using the World Checklist of Vascular Plants as a reference, we generate random samples of species with increasing sample sizes from the global species pool. We compare the diversity patterns retrieved from the samples against the patterns of the reference dataset using spatially weighted correlation coefficients and four different diversity metrics. We find that at the botanical country scale, representative global maps of species and phylogenetic diversity can be created with small numbers of species (~1% [0.2% and 0.4%, respectively]) at the botanical country scale. For effective growth form and family diversity sample sizes encompassing ~20% [19.2% and 19.5%, respectively] of all species are needed. Random samples require markedly fewer species to reach high correlations than when restricting the pool of species to single plant families or genera. We show that when representative samples are used robust inferences of plant diversity patterns can be made from only a small proportion of species.

The shift from outcrossing to self-fertilization is a common evolutionary transition in flowering plants. The genus Capsella, comprising the obligate outcrosser C. grandiflora and two self-fertile species, C. rubella and C. orientalis, provides a powerful system to explore genomic consequences of mating system shifts. Despite its utility, existing genomic resources in Capsella are fragmented, incomplete, and particularly deficient in repetitive genomic regions, hindering the study of transposable element (TE) dynamics and gene annotation. Here, we present high-quality, chromosome-scale, near-gapless genome assemblies for C. grandiflora, C. rubella, and C. orientalis. Leveraging these improved genomes, we created high-quality genomic resources for the Capsella genus by performing comprehensive, de novo annotations of protein-coding genes and TEs. Comparative genomic analysis among these species reveals differences in TE abundance, position, and production of small RNAs. These resources provide an unprecedented opportunity to explore how mating system transitions influence genome architecture, TE behavior, and gene evolution. This research also developed a static online platform for Capsella genomic resources, Capsella Database (CapBase, www.capsella.uk), to support community use of these resources. Our findings advance understanding of the genomic impacts of selfing and establish a robust foundation for future research into genomics, epigenomics, and evolutionary biology within Capsella and related plant systems.

O_LITerrestrial biosphere models commonly use empirical scaling factors to represent soil moisture constraints on carbon and water fluxes, but these lack mechanistic grounding and produce inconsistent estimates of soil moisture limitations on primary production and transpiration across models. C_LIO_LIHere we extended the least-cost hypothesis for optimal stomatal conductance to account for soil moisture limitations by allowing soil water availability to modulate the carbon cost of water transport, drawing on the observed temperature dependence of stem respiration, and derived a simple empirical approximation to the theory using global {delta}13C and eddy covariance data. C_LIO_LIThe empirical analysis shows moderated thermal acclimation of stem respiration and a weak increase in water transport costs with aridity, supporting the interpretation that the decline in light-use efficiency (LUE) under arid conditions is primarily attributable to non-stomatal limitations. C_LIO_LIValidation against an independent global dataset of sapflow-derived canopy conductance and transpiration shows that the revised scheme significantly improves the predictive power of the least-cost hypothesis, offering a more mechanistically coherent alternative to existing soil moisture parameterisations. C_LI