Improvement on the genetic engineering of an invasive agricultural pest insect, the cherry vinegar fly, Drosophila suzukii

EResNet-SVM: an overfitting-relieved deep learning model for recognition of plant diseases and pests

BACKGROUND

The accurate recognition and early warning for plant diseases and pests are a prerequisite of intelligent prevention and control for plant diseases and pests. As a result of the phenotype similarity of the hazarded plant after plant diseases and pests occur, as well as the interference of the external environment, traditional deep learning models often face the overfitting problem in phenotype recognition of plant diseases and pests, which leads to not only the slow convergence speed of the network, but also low recognition accuracy.

RESULTS

Motivated by the above problems, the present study proposes a deep learning model EResNet-support vector machine (SVM) to alleviate the overfitting for the recognition and classification of plant diseases and pests. First, the feature extraction capability of the model is improved by increasing feature extraction layers in the convolutional neural network. Second, the order-reduced modules are embedded and a sparsely activated function is introduced to reduce model complexity and alleviate overfitting. Finally, a classifier fused by SVM and fully connected layers are introduced to transforms the original non-linear classification problem into a linear classification problem in high-dimensional space to further alleviate the overfitting and improve the recognition accuracy of plant diseases and pests. The ablation experiments further demonstrate that the fused structure can effectively alleviate the overfitting and improve the recognition accuracy. The experimental recognition results for typical plant diseases and pests show that the proposed EResNet-SVM model has 99.30% test accuracy for eight conditions (seven plant diseases and one normal), which is 5.90% higher than the original ResNet18. Compared with the classic AlexNet, GoogLeNet, Xception, SqueezeNet and DenseNet201 models, the accuracy of the EResNet-SVM model has improved by 5.10%, 7%, 8.10%, 6.20% and 1.90%, respectively. The testing accuracy of the EResNet-SVM model for 6 insect pests is 100%, which is 3.90% higher than that of the original ResNet18 model.

CONCLUSION

This research provides not only useful references for alleviating the overfitting problem in deep learning, but also a theoretical and technical support for the intelligent detection and control of plant diseases and pests. © 2024 Society of Chemical Industry.

Protocol for genetic engineering in Drosophila suzukii using microinjection

The spotted wing Drosophila (Drosophila suzukii Matsumura) is recognized globally as a significant economic pest. Here, we present a protocol for genetic engineering in D. suzukii using microinjection. We describe steps for genetic engineering techniques, including transposon-mediated germline transformation, recombinase-mediated genome targeting, and CRISPR-mediated gene editing. This protocol can significantly expand the toolkit for functional genomics and genetic control studies of this pest. For complete details on the use and execution of this protocol, please refer to Schetelig and Handler,1 Schetelig et al.2 Yan et al.,3 and Yan et al.4.

Expansion of the genetic toolbox for manipulation of the global crop pest Drosophila suzukii: Isolation and assessment of eye colour mutant strains

Drosophila suzukii (Matsumura) (Diptera: Drosophilidae), commonly called spotted wing Drosophila, is an important agricultural pest recognised worldwide. D. suzukii is a pest of soft-skinned fruits as females can lay eggs in ripening fruit before harvest. While strains for genetic biocontrol of D. suzukii have been made, the development of transgenic D. suzukii strains and their further screening remain a challenge partly due to the lack of phenotypically trackable genetic-markers, such as those widely used with the model genetic organism D. melanogaster. Here, we have used CRISPR/Cas9 to introduce heritable mutations in the eye colour genes white, cinnabar and sepia, which are located on the X, second and third chromosomes, respectively. Strains were obtained, which were homozygous for a single mutation. Genotyping of the established strains showed insertion and/or deletions (indels) at the targeted sites. A strain homozygous for mutations in cinnabar and sepia showed a pale-yellow eye colour at eclosion but darkened to a sepia colour after a week. The fecundity and fertility of some of the cinnabar and sepia strains were comparable with the wild type. Although white mutant males were previously reported to be sterile, we found that sterility is not fully penetrant and we have been able to maintain white-eyed strains for over a year. The cinnabar, sepia and white mutant strains developed in this study should facilitate future genetic studies in D. suzukii and the development of strains for genetic control of this pest.

Improved piggyBac Transformation with Capped Transposase mRNA in Pest Insects

Creating transgenic insects is a key technology in insect genetics and molecular biology. A widely used instrument in insect transgenesis is the piggyBac transposase, resulting in essentially random genomic integrations. In contrast, site-specific recombinases allow the targeted integration of the transgene construct into a specific genomic target site. Both strategies, however, often face limitations due to low transgenesis efficiencies. We aimed to enhance transgenesis efficiencies by utilizing capped mRNA as a source of transposase or recombinase instead of a helper plasmid. A systematic comparison of transgenesis efficiencies in Aedes mosquitoes, as models for hard-to-transform insects, showed that suppling piggyBac transposase as mRNA increased the average transformation efficiency in Aedes aegypti from less than 5% with the plasmid source to about 50% with mRNA. Similar high activity was observed in Ae. albopictus with pBac mRNA. No efficiency differences between plasmid and mRNA were observed in recombination experiments. Furthermore, a hyperactive version of piggyBac transposase delivered as a plasmid did not improve the transformation efficiency in Ae. aegypti or the agricultural pest Drosophila suzukii. We believe that the use of mRNA has strong potential for enhancing piggyBac transformation efficiencies in other mosquitoes and important agricultural pests, such as tephritids.

Genetically Encoded CRISPR Components Yield Efficient Gene Editing in the Invasive Pest Drosophila suzukii

Originally from Asia, Drosophila suzukii Matsumura is a global pest of economically important soft-skinned fruits. Also commonly known as spotted wing drosophila, it is largely controlled through repeated applications of broad-spectrum insecticides by which resistance has been observed in the field. There is a pressing need for a better understanding of D. suzukii biology and for developing alternative environmentally friendly methods of control. The RNA-guided Cas9 nuclease has revolutionized functional genomics and is an integral component of several recently developed genetic strategies for population control of insects. Here, we describe genetically modified strains that encode three different terminators and four different promoters to express Cas9 robustly in both the soma and/or germline of D. suzukii. The Cas9 strains were rigorously evaluated through genetic crossing to transgenic strains that encode single-guide RNAs targeting the conserved X-linked yellow body and white eye genes. We find that several Cas9/gRNA strains display remarkably high editing capacity. Going forward, these tools will be instrumental for evaluating gene function in D. suzukii and may even provide tools useful for the development of new genetic strategies for control of this invasive species.