HairNet2: deep learning to quantify cotton leaf hairiness, a complex genetic and environmental trait.

BACKGROUND: Cotton accounts for 80% of the global natural fibre production. Its leaf hairiness affects insect resistance, fibre yield, and economic value. However, this phenotype is still qualitatively assessed by visually attributing a Genotype Hairiness Score (GHS) to a leaf/plant, or by using the HairNet deep-learning model which also outputs a GHS. Here, we introduce HairNet2, a quantitative deep-learning model which detects leaf hairs (trichomes) from images and outputs a segmentation mask and a Leaf Trichome Score (LTS). RESULTS: Trichomes of 1250 images were annotated (AnnCoT) and a combination of six Feature Extractor modules and five Segmentation modules were tested alongside a range of loss functions and data augmentation techniques. HairNet2 was further validated on the dataset used to build HairNet (CotLeaf-1), a similar dataset collected in two subsequent seasons (CotLeaf-2), and a dataset collected on two genetically diverse populations (CotLeaf-X). The main findings of this study are that (1) leaf number, environment and image position did not significantly affect results, (2) although GHS and LTS mostly correlated for individual GHS classes, results at the genotype level revealed a strong LTS heterogeneity within a given GHS class, (3) LTS correlated strongly with expert scoring of individual images. CONCLUSIONS: HairNet2 is the first quantitative and scalable deep-learning model able to measure leaf hairiness. Results obtained with HairNet2 concur with the qualitative values used by breeders at both extremes of the scale (GHS 1-2, and 5-5+), but interestingly suggest a reordering of genotypes with intermediate values (GHS 3-4+). Finely ranking mild phenotypes is a difficult task for humans. In addition to providing assistance with this task, HairNet2 opens the door to selecting plants with specific leaf hairiness characteristics which may be associated with other beneficial traits to deliver better varieties.

The importance of species-specific and temperature-sensitive parameterisation of A/Ci models: A case study using cotton (Gossypium hirsutum L.) and the automated 'OptiFitACi' R-package.

Leaf gas exchange measurements are an important tool for inferring a plant's photosynthetic biochemistry. In most cases, the responses of photosynthetic CO2 assimilation to variable intercellular CO2 concentrations (A/Ci response curves) are used to model the maximum (potential) rate of carboxylation by ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco, Vcmax) and the rate of photosynthetic electron transport at a given incident photosynthetically active radiation flux density (PAR; JPAR). The standard Farquhar-von Caemmerer-Berry model is often used with default parameters of Rubisco kinetic values and mesophyll conductance to CO2 (gm) derived from tobacco that may be inapplicable across species. To study the significance of using such parameters for other species, here we measured the temperature responses of key in vitro Rubisco catalytic properties and gm in cotton (Gossypium hirsutum cv. Sicot 71) and derived Vcmax and J2000 (JPAR at 2000 µmol m-2 s-1 PAR) from cotton A/Ci curves incrementally measured at 15°C-40°C using cotton and other species-specific sets of input parameters with our new automated fitting R package 'OptiFitACi'. Notably, parameterisation by a set of tobacco parameters produced unrealistic J2000:Vcmax ratio of <1 at 25°C, two- to three-fold higher estimates of Vcmax above 15°C, up to 2.3-fold higher estimates of J2000 and more variable estimates of Vcmax and J2000, for our cotton data compared to model parameterisation with cotton-derived values. We determined that errors arise when using a gm,25 of 2.3 mol m-2 s-1 MPa-1 or less and Rubisco CO2-affinities in 21% O2 (KC 21%O2) at 25°C outside the range of 46-63 Pa to model A/Ci responses in cotton. We show how the A/Ci modelling capabilities of 'OptiFitACi' serves as a robust, user-friendly, and flexible extension of 'plantecophys' by providing simplified temperature-sensitivity and species-specificity parameterisation capabilities to reduce variability when modelling Vcmax and J2000.

Response of the plant core microbiome to Fusarium oxysporum infection and identification of the pathobiome.

Plant core microbiomes consist of persistent key members that provide critical host functions, but their assemblages can be interrupted by biotic and abiotic stresses. The pathobiome is comprised of dynamic microbial interactions in response to disease status of the host. Hence, identifying variation in the core microbiome and pathobiome can significantly advance our understanding of microbial-microbial interactions and consequences for disease progression and host functions. In this study, we combined glasshouse and field studies to analyse the soil and plant rhizosphere microbiome of cotton plants (Gossypium hirsutum) in the presence of a cotton-specific fungal pathogen, Fusarium oxysporum f. sp. vasinfectum (FOV). We found that FOV directly and consistently altered the rhizosphere microbiome, but the biocontrol agents enabled microbial assemblages to resist pathogenic stress. Using co-occurrence network analysis of the core microbiome, we identified the pathobiome comprised of the pathogen and key associate phylotypes in the cotton microbiome. Isolation and application of some negatively correlated pathobiome members provided protection against plant infection. Importantly, our field survey from multiple cotton fields validated the pattern and responses of core microbiomes under FOV infection. This study advances key understanding of core microbiome responses and existence of plant pathobiomes, which provides a novel framework to better manage plant diseases in agriculture and natural settings.

Cotton Breeding in Australia: Meeting the Challenges of the 21st Century.

The Commonwealth Scientific and Industrial Research Organisation (CSIRO) cotton breeding program is the sole breeding effort for cotton in Australia, developing high performing cultivars for the local industry which is worth∼AU$3 billion per annum. The program is supported by Cotton Breeding Australia, a Joint Venture between CSIRO and the program's commercial partner, Cotton Seed Distributors Ltd. (CSD). While the Australian industry is the focus, CSIRO cultivars have global impact in North America, South America, and Europe. The program is unique compared with many other public and commercial breeding programs because it focuses on diverse and integrated research with commercial outcomes. It represents the full research pipeline, supporting extensive long-term fundamental molecular research; native and genetically modified (GM) trait development; germplasm enhancement focused on yield and fiber quality improvements; integration of third-party GM traits; all culminating in the release of new commercial cultivars. This review presents evidence of past breeding successes and outlines current breeding efforts, in the areas of yield and fiber quality improvement, as well as the development of germplasm that is resistant to pests, diseases and abiotic stressors. The success of the program is based on the development of superior germplasm largely through field phenotyping, together with strong commercial partnerships with CSD and Bayer CropScience. These relationships assist in having a shared focus and ensuring commercial impact is maintained, while also providing access to markets, traits, and technology. The historical successes, current foci and future requirements of the CSIRO cotton breeding program have been used to develop a framework designed to augment our breeding system for the future. This will focus on utilizing emerging technologies from the genome to phenome, as well as a panomics approach with data management and integration to develop, test and incorporate new technologies into a breeding program. In addition to streamlining the breeding pipeline for increased genetic gain, this technology will increase the speed of trait and marker identification for use in genome editing, genomic selection and molecular assisted breeding, ultimately producing novel germplasm that will meet the coming challenges of the 21st Century.

Genomic prediction of cotton fibre quality and yield traits using Bayesian regression methods.

Genomic selection or genomic prediction (GP) has increasingly become an important molecular breeding technology for crop improvement. GP aims to utilise genome-wide marker data to predict genomic breeding value for traits of economic importance. Though GP studies have been widely conducted in various crop species such as wheat and maize, its application in cotton, an essential renewable textile fibre crop, is still significantly underdeveloped. We aim to develop a new GP-based breeding system that can improve the efficiency of our cotton breeding program. This article presents a GP study on cotton fibre quality and yield traits using 1385 breeding lines from the Commonwealth Scientific and Industrial Research Organisation (CSIRO, Australia) cotton breeding program which were genotyped using a high-density SNP chip that generated 12,296 informative SNPs. The aim of this study was twofold: (1) to identify the models and data sources (i.e. genomic and pedigree) that produce the highest prediction accuracies; and (2) to assess the effectiveness of GP as a selection tool in the CSIRO cotton breeding program. The prediction analyses were conducted under various scenarios using different Bayesian predictive models. Results highlighted that the model combining genomic and pedigree information resulted in the best cross validated prediction accuracies: 0.76 for fibre length, 0.65 for fibre strength, and 0.64 for lint yield. Overall, this work represents the largest scale genomic selection studies based on cotton breeding trial data. Prediction accuracies reported in our study indicate the potential of GP as a breeding tool for cotton. The study highlighted the importance of incorporating pedigree and environmental factors in GP models to optimise the prediction performance.

Core Collections: Is There Any Value for Cotton Breeding?

Global plant breeding activities are reliant on the available genetic variation held in extant varieties and germplasm collections. Throughout the mid- to late 1900s, germplasm collecting efforts were prioritized for breeding programs to archive precious material before it disappeared and led to the development of the numerous large germplasm resources now available in different countries. In recent decades, however, the maintenance and particularly the expansion of these germplasm resources have come under threat, and there has been a significant decline in investment in further collecting expeditions, an increase in global biosecurity restrictions, and restrictions placed on the open exchange of some commercial germplasm between breeders. The large size of most genebank collections, as well as constraints surrounding the availability and reliability of accurate germplasm passport data and physical or genetic characterization of the accessions in collections, limits germplasm utilization by plant breeders. To overcome these constraints, core collections, defined as a representative subset of the total germplasm collection, have gained popularity. Core collections aim to increase germplasm utilization by containing highly characterized germplasm that attempts to capture the majority of the variation in a whole collection. With the recent availability of many new genetic tools, the potential to unlock the value of these resources can now be realized. The Commonwealth Scientific and Industrial Research Organisation (CSIRO) cotton breeding program supplies 100% of the cotton cultivars grown in Australia. The program is reliant on the use of plant genetic resources for the development of improved cotton varieties to address emerging challenges in pest and disease resistance as well as the global changes occurring in the climate. Currently, the CSIRO germplasm collection is actively maintained but underutilized by plant breeders. This review presents an overview of the Australian cotton germplasm resources and discusses the appropriateness of a core collection for cotton breeding programs.

Understanding and Exploiting Transpiration Response to Vapor Pressure Deficit for Water Limited Environments.

More frequent droughts and an increased pressure on water resources, combined with social licence to operate, will inevitably decrease water resources available for fully irrigated cotton production. Therefore, the long-term future of the cotton industry will require more drought tolerant varieties that can perform well when grown in rainfed cropping regions often exposed to intermittent drought. A trait that limits transpiration (TRLim) under an increased vapour pressure deficit (VPD) may increase crop yield in drier atmospheric conditions and potentially conserve soil water to support crop growth later in the growing season. However, this trait has not been tested or identified in cotton production systems. This study tested the hypotheses that (1) genetic variability to the TRLim VPD trait exists amongst 10 genotypes in the Australian cotton breeding programme; (2) genotypes with a TRLim VPD trait use less water in high VPD environments and (3) variation in yield responses of cotton genotypes is linked with the VPD environment and water availability during the peak flowering period. This study combined glasshouse and field experiments to assess plant transpiration and crop yield responses of predominantly locally bred cotton genotypes to a range of atmospheric VPD under Australian climatic conditions. Results indicated that genetic variation to the limiting transpiration VPD trait exists within cotton genotypes in the Australian breeding programme, with five genotypes identified as expressing the TRLim VPD trait. A modelling study suggests that this trait may not necessarily result in overall reduced plant water use due to greater transpiration rates at lower VPD environments negating the water conservation in high VPD environments. However, our study showed that the yield response of cotton genotypes is linked with both VPD environment and water availability during the peak flowering period. Yield performance of the TRLim genotype was improved at some high VPD environments but is unlikely to out-perform a genotype with a lower yield potential. Improved understanding of integrated plant- and crop-level genotypic responses to the VPD environments will enhance germplasm development to benefit cotton production in both rainfed and semi-irrigated cotton systems, thereby meeting the agricultural challenges of the twenty-first Century.

HairNet: a deep learning model to score leaf hairiness, a key phenotype for cotton fibre yield, value and insect resistance.

BACKGROUND: Leaf hairiness (pubescence) is an important plant phenotype which regulates leaf transpiration, affects sunlight penetration, and provides increased resistance or susceptibility against certain insects. Cotton accounts for 80% of global natural fibre production, and in this crop leaf hairiness also affects fibre yield and value. Currently, this key phenotype is measured visually which is slow, laborious and operator-biased. Here, we propose a simple, high-throughput and low-cost imaging method combined with a deep-learning model, HairNet, to classify leaf images with great accuracy. RESULTS: A dataset of [Formula: see text] 13,600 leaf images from 27 genotypes of Cotton was generated. Images were collected from leaves at two different positions in the canopy (leaf 3 & leaf 4), from genotypes grown in two consecutive years and in two growth environments (glasshouse & field). This dataset was used to build a 4-part deep learning model called HairNet. On the whole dataset, HairNet achieved accuracies of 89% per image and 95% per leaf. The impact of leaf selection, year and environment on HairNet accuracy was then investigated using subsets of the whole dataset. It was found that as long as examples of the year and environment tested were present in the training population, HairNet achieved very high accuracy per image (86-96%) and per leaf (90-99%). Leaf selection had no effect on HairNet accuracy, making it a robust model. CONCLUSIONS: HairNet classifies images of cotton leaves according to their hairiness with very high accuracy. The simple imaging methodology presented in this study and the high accuracy on a single image per leaf achieved by HairNet demonstrates that it is implementable at scale. We propose that HairNet replaces the current visual scoring of this trait. The HairNet code and dataset can be used as a baseline to measure this trait in other species or to score other microscopic but important phenotypes.

A standardised approach for determining heat tolerance in cotton using triphenyl tetrazolium chloride.

Improving the heat tolerance of cotton is a major concern for breeding programs. To address this need, a fast and effect way of quantifying thermotolerant phenotypes is required. Triphenyl tetrazolium chloride (TTC) based enzyme viability testing following high-temperature stress can be used as a vegetative heat tolerance phenotype. This is because when live cells encounter a TTC solution, TTC undergoes a chemical reduction producing a visible, insoluble red product called triphenyl formazan, that can be quantified spectrophotometrically. However, existing TTC based cell viability assays cannot easily be deployed at the scale required in a crop improvement program. In this study, a heat stress assay (HSA) based on the use of TTC enzyme viability testing has been refined and improved for efficiency, reliability, and ease of use through four experiments. Sampling factors that may influence assay results, such as leaf age, plant water status, and short-term cold storage, were also investigated. Experiments conducted in this study have successfully downscaled the assay and identified an optimal sampling regime, enabling measurement of large segregating populations for application in breeding programs. The improved HSA methodology is important as it is proposed that long-term improvements in cotton thermotolerance can be achieved through the concurrent selection of superior phenotypes based on the HSA and yield performance in hot environments. Additionally, a new way of interpreting both heat tolerance and heat resistance was developed, differentiating genotypes that perform well at the time of a heat stress event and those that maintain a similar performance level to a non-stressed control.

Historical Datasets Support Genomic Selection Models for the Prediction of Cotton Fiber Quality Phenotypes Across Multiple Environments.

Genomic selection (GS) has successfully been used in plant breeding to improve selection efficiency and reduce breeding time and cost. However, there has not been a study to evaluate GS prediction models that may be used for predicting cotton breeding lines across multiple environments. In this study, we evaluated the performance of Bayes Ridge Regression, BayesA, BayesB, BayesC and Reproducing Kernel Hilbert Spaces regression models. We then extended the single-site GS model to accommodate genotype × environment interaction (G×E) in order to assess the merits of multi- over single-environment models in a practical breeding and selection context in cotton, a crop for which this has not previously been evaluated. Our study was based on a population of 215 upland cotton (Gossypium hirsutum) breeding lines which were evaluated for fiber length and strength at multiple locations in Australia and genotyped with 13,330 single nucleotide polymorphic (SNP) markers. BayesB, which assumes unique variance for each marker and a proportion of markers to have large effects, while most other markers have zero effect, was the preferred model. GS accuracy for fiber length based on a single-site model varied across sites, ranging from 0.27 to 0.77 (mean = 0.38), while that of fiber strength ranged from 0.19 to 0.58 (mean = 0.35) using randomly selected sub-populations as the training population. Prediction accuracies from the M×E model were higher than those for single-site and across-site models, with an average accuracy of 0.71 and 0.59 for fiber length and strength, respectively. The use of the M×E model could therefore identify which breeding lines have effects that are stable across environments and which ones are responsible for G×E and so reduce the amount of phenotypic screening required in cotton breeding programs to identify adaptable genotypes.

Vapour pressure deficit aids the interpretation of cotton canopy temperature response to water deficit.

Crop canopy temperature (Tc) is coupled with transpiration, which is a function of soil and atmospheric conditions and plant water status. Thus, Tc has been identified as a real-time, plant-based tool for crop water stress detection. Such plant-based methods theoretically integrate the water status of both the plant and its environment. However, previous studies have highlighted the limitations and difficulty of interpreting the Tc response to plant and soil water stress. This study investigates the links between cotton Tc, established measures of plant water relations and atmospheric vapour pressure deficit (VPDa). Concurrent measures of carbon assimilation (A), stomatal conductance (gs), leaf water potential (Ψl), soil water (fraction of transpirable soil water (FTSW)) and Tc were conducted in surface drip irrigated cotton over two growing seasons. Associations between A, gs, Ψl, FTSW and Tc are presented, which are significantly improved with the inclusion of VPDa. It was concluded that utilising the strong associations between Ψl, VPDa and Tc, an adjustment of 1.8°C for each unit of VPDa should be made to the critical Tc for irrigation. This will improve the precision of irrigation in Tc based irrigation scheduling protocols. Improved accuracy in water stress detection with Tc, and an understanding of the interaction the environment plays in this response, can potentially improve the efficiency of irrigation.