Urban landscapes are often characterized by a wide range of diverse flowering plants consisting of native and exotic plants. These flower-rich habitats have proven to be particularly valuable for urban pollinating insects. However, the ability of ornamental plants in supporting urban pollinator communities is still not well documented. For this study, we established flower beds at 13 different urban testing sites, which were planted with identical sets of ornamental garden plants. The pollinator visitation patterns were then observed throughout the summer seasons. Over a two-year period, a total of 10,565 pollinators were recorded with wild bees (> 50%, excluding bumblebees) being the most abundant pollinator group. Our results revealed that (I) the assortment of ornamental plants was visited by a high number of urban pollinators for collecting pollen and nectar, and (II) the pollinator abundance and composition varied significantly within the tested ornamental plants. These differences occurred not only among plant species but to the same extent among cultivars, whereby the number of pollinators was positively correlated with number of flowering units per plant. By using a generalized linear mixed model (GLMM) and redundancy analysis (RDA) we identified further significant impacts of the two variables year and location on the insect pollinator abundance and richness. Despite of the local and yearly variations, our approach provided a good and field-applicable method to evaluate the pollinator friendliness in ornamental plants. Such tools are urgently required to validate labels like ‘bee friendly’ or ‘pollinator friendly’ used by plant breeding companies.
Background Maize cobs are an important component of crop yield that exhibit a high diversity in size, shape and color in native landraces and modern varieties. Various phenotyping approaches were developed to measure maize cob parameters in a high throughput fashion. More recently, deep learning methods like convolutional neural networks (CNNs) became available and were shown to be highly useful for high-throughput plant phenotyping. We aimed at comparing classical image segmentation with deep learning methods for maize cob image segmentation and phenotyping using a large image dataset of native maize landrace diversity from Peru. Results Comparison of three image analysis methods showed that a Mask R-CNN trained on a diverse set of maize cob images was highly superior to classical image analysis using the Felzenszwalb-Huttenlocher algorithm and a Window-based CNN due to its robustness to image quality and object segmentation accuracy ($$r=0.99$$ r = 0.99 ). We integrated Mask R-CNN into a high-throughput pipeline to segment both maize cobs and rulers in images and perform an automated quantitative analysis of eight phenotypic traits, including diameter, length, ellipticity, asymmetry, aspect ratio and average values of red, green and blue color channels for cob color. Statistical analysis identified key training parameters for efficient iterative model updating. We also show that a small number of 10–20 images is sufficient to update the initial Mask R-CNN model to process new types of cob images. To demonstrate an application of the pipeline we analyzed phenotypic variation in 19,867 maize cobs extracted from 3449 images of 2484 accessions from the maize genebank of Peru to identify phenotypically homogeneous and heterogeneous genebank accessions using multivariate clustering. Conclusions Single Mask R-CNN model and associated analysis pipeline are widely applicable tools for maize cob phenotyping in contexts like genebank phenomics or plant breeding.
Background: Maize cobs are an important component of crop yield that exhibit a high diversity in size, shape and color in native landraces and modern varieties. Various phenotyping approaches were developed to measure maize cob parameters in a high throughput fashion. More recently, deep learning methods like convolutional neural networks (CNN) became available and were shown to be highly useful for high-throughput plant phenotyping. We aimed at comparing classical image segmentation with deep learning methods for maize cob image segmentation and phenotyping using a large image dataset of native maize landrace diversity from Peru. Results: Comparison of three image analysis methods showed that a Mask R-CNN trained on a diverse set of maize cob images was highly superior to classical image analysis using the Felzenszwalb-Huttenlocher algorithm and a Window-based CNN due to its robustness to image quality and object segmentation accuracy (r=0.99). We integrated Mask R-CNN into a high-throughput pipeline to segment both maize cobs and rulers in images and perform an automated quantitative analysis of eight phenotypic traits, including diameter, length, ellipticity, asymmetry, aspect ratio and average RGB values for cob color. Statistical analysis identified key training parameters for efficient iterative model updating. We also show that a small number of 10-20 images is sufficient to update the initial Mask R-CNN model to process new types of cob images. To demonstrate an application of the pipeline we analyzed phenotypic variation in 19,867 maize cobs extracted from 3,449 images of 2,484 accessions from the maize genebank of Peru to identify phenotypically homogeneous and heterogeneous genebank accessions using multivariate clustering. Conclusions: Single Mask R-CNN model and associated analysis pipeline are widely applicable tools for maize cob phenotyping in contexts like genebank phenomics or plant breeding.
Ornamental plants are appreciated by humans for their colorfulness, beauty, abundant flowering and long blooming periods. Many ornamental plants can also constitute an additional foraging resource for flower-visiting insects. However, the ability of the popular ornamental plant Calibrachoa to support urban insect communities is not well documented. In this study, 20 different Calibrachoa cultivars were selected and tested in regard to their insect friendliness based on standardized observations (I) in flight tents using the large earth bumble bee Bombus terrestris as a model species and (II) in open field trials. To investigate what floral characteristics might constitute attractiveness to bumble bees, various floral traits were recorded and compared across all tested Calibrachoa cultivars. Over a two-year period, a total of 6,327 foraging bumble bees were recorded in the tent observations. In the open field observations, we counted 4,188 flower-visiting insects. Our results revealed that (I) all Calibrachoa cultivars were visited by insects for foraging, (II) the number of insect visitors varied significantly among the 20 tested cultivars and (III) the cultivars displayed different floral traits. For the morphometric floral traits and the aroma profiles of Calibrachoa, only the mean nectar quantity and a few identified compounds could be correlated with attractiveness to the model species B. terrestris. We also found that the petal color of the tested cultivars had a significant impact on the number of visitors. Therefore, B. terrestris clearly preferred red or blue Calibrachoa cultivars over those with other petal colors. However, as the cultivar preferences in the different insect groups differed, it is highly recommended to use various cultivars in urban plantings. Nevertheless, efforts must be made to explain what additional floral traits make Calibrachoa and other ornamental plants generally attractive to flower visitors. This information can then be used for breeding purposes to increase the insect friendliness of ornamental plants.
Powdery mildew caused by Podosphaera xanthii is among the most threatening fungal diseases affecting melons on the Mediterranean coast. Although the use of genetic resistance is a highly recommended alternative to control this pathogen, many races of this fungus have been described and, therefore, resistance is usually overcome; thus, breeding for resistance to this pathogen is a challenge. Several melon genotypes carrying resistance to powdery mildew have been described but their agronomical and fruit characters are usually far away from the required melon types in many commercial markets. Taking this into consideration, looking for novel sources of resistance in Tunisian landraces is a very convenient step to obtain new resistant melon varieties/hybrids suitable for Mediterranean markets. Several Tunisian melon landraces have been tested against three common races in Mediterranean regions (Race 2, Race 3.5, and Race 5), using phenotypic approaches in two independent experiments (artificial inoculations in a growth chamber and natural conditions of infection in a greenhouse). The results of the artificial inoculations showed that all the tested landraces were susceptible to Race 3.5 and Race 5 and several landraces were resistant to Race 2. Under natural conditions of infection, Race 2 of P. xanthii was the race prevalent in the plot and the resistance of TUN-16, TUN-19, and TUN-25 was confirmed. The found resistances were race-specific and underlie a high genetic influence reflected in the high value of the estimated heritability of 0.86. These resistant landraces should be considered as a potential source of resistance in breeding programs of melons belonging to inodorus and reticulatus groups, but further research is necessary to elucidate the genetic control of the found resistances and to provide useful molecular markers linked to P. xanthii Race 2 resistance.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.