Explore the vast range of titles available.

A small sample size and unbalanced sample distribution are two main problems when data-driven methods are applied for fault diagnosis in practical engineering. Technically, sample generation and data augmentation have proven to be effective methods to solve this problem. The generative adversarial network (GAN) has been widely used in recent years as a representative generative model. Besides the general GAN, many variants have recently been reported to address its inherent problems such as mode collapse and slow convergence. In addition, many new techniques are being proposed to increase the sample generation quality. Therefore, a systematic review of GAN, especially its application in fault diagnosis, is necessary. In this paper, the theory and structure of GAN and variants such as ACGAN, VAEGAN, DCGAN, WGAN, et al. are presented first. Then, the literature on GANs is mainly categorized and analyzed from two aspects: improvements in GAN’s structure and loss function. Specifically, the improvements in the structure are classified into three types: information-based, input-based, and layer-based. Regarding the modification of the loss function, it is sorted into two aspects: metric-based and regularization-based. Afterwards, the evaluation metrics of the generated samples are summarized and compared. Finally, the typical applications of GAN in the bearing fault diagnosis field are listed, and the challenges for further research are also discussed.

The development of state estimators for local electrical energy supply systems is inevitable as the role of the system’s become more important, especially with the recent increased interest in direct current (DC) microgrids. Proper control and monitoring requires a state estimator that can adapt to the new technologies for DC microgrids. This paper mainly deals with the DC microgrid state estimation (SE) using a modified long short-term memory (LSTM) network, which until recently has been applied only in forecasting studies. The modified LSTM network for the proposed state estimator adopted a specifically weighted least square (WLS)-based loss function for training. To demonstrate the performance of the proposed state estimator, a comparison study was done with other SE methods included in this paper. The results showed that the proposed state estimator had high accuracy in estimating the states of DC microgrids. Other than the enhanced accuracy, the deep-learning-based state estimator also provided faster computation speeds than the conventional state estimator.

NRC publication: Yes

This work is devoted to improving the quality of segmentation of images of various scientific papers and legal acts by neural network models by training them using modified loss functions that take into account special features of images of the appropriate subject domain. The analysis of existing loss functions is carried out, and new functions are proposed that work both with the coordinates of bounding boxes and use information about the pixels of the input image. To assess the quality, a neural network segmentation model with modified loss functions is trained, and a theoretical assessment is carried out using a simulation experiment showing the convergence rate and segmentation error. As a result of the study, rapidly converging loss functions are created that improve the quality of document image segmentation using additional information about the input data.

Gene-editing technologies, which include the CRISPR–Cas nucleases 1 – 3 and CRISPR base editors 4 , 5 , have the potential to permanently modify disease-causing genes in patients 6 . The demonstration of durable editing in target organs of nonhuman primates is a key step before in vivo administration of gene editors to patients in clinical trials. Here we demonstrate that CRISPR base editors that are delivered in vivo using lipid nanoparticles can efficiently and precisely modify disease-related genes in living cynomolgus monkeys ( Macaca fascicularis ). We observed a near-complete knockdown of PCSK9 in the liver after a single infusion of lipid nanoparticles, with concomitant reductions in blood levels of PCSK9 and low-density lipoprotein cholesterol of approximately 90% and about 60%, respectively; all of these changes remained stable for at least 8 months after a single-dose treatment. In addition to supporting a ‘once-and-done’ approach to the reduction of low-density lipoprotein cholesterol and the treatment of atherosclerotic cardiovascular disease (the leading cause of death worldwide 7 ), our results provide a proof-of-concept for how CRISPR base editors can be productively applied to make precise single-nucleotide changes in therapeutic target genes in the liver, and potentially in other organs. In a cynomolgus macaque model, CRISPR base editors delivered in lipid nanoparticles are shown to efficiently and stably knock down PCSK9 in the liver to reduce levels of PCSK9 and low-density lipoprotein cholesterol in the blood.

Key message CRISPR/Cas9-mediated editing of Clpsk1 enhanced watermelon resistance to Fusarium oxysporum . The clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 system has proven to be an effective genome-editing tool for crop improvement. Previous studies described that Phytosulfokine (PSK) signalling attenuates plant immune response. In this work, we employed the CRISPR/Cas9 system to knockout Clpsk1 gene, encoding the PSK precursor, to confer enhanced watermelon resistance to Fusarium oxysporum f.sp. niveum ( FON ). Interactions between PSK and FON were analysed and it was found that transcript of Clpsk1 was significantly induced upon FON infection. Meanwhile, application of exogenous PSK increased the pathogen growth. Then, one sgRNA, which targeted the first exon of Clpsk1 , was selected for construction of pRGEB32-CAS9-gRNA-Clpsk1 expression cassette. The construct was then transformed to watermelon through Agrobacterium tumefaciens -mediated transformation method. Six mutant plants were obtained and three types of mutations at the expected position were identified based on Sanger sequencing. Resistance evaluation indicated that Clpsk1 loss-of-function rendered watermelon seedlings more resistant to infection by FON . These results indicate that CRISPR/Cas9-mediated gene modification is an effective approach for watermelon improvement.

• Seed weight is one of the most important agronomic traits in soybean for yield improvement and food production. Several quantitative trait loci (QTLs) associated with the trait have been identified in soybean. However, the genes underlying the QTLs and their functions remain largely unknown. • Using forward genetic methods and CRISPR/Cas9 gene editing, we identified and characterized the role of GmKIX8-1 in the control of organ size in soybean. • GmKIX8-1 belongs to a family of KIX domain-containing proteins that negatively regulate cell proliferation in plants. Consistent with this predicted function, we found that loss-of-function GmKIX8-1 mutants showed a significant increase in the size of aerial plant organs, such as seeds and leaves. Likewise, the increase in organ size is due to increased cell proliferation, rather than cell expansion, and increased expression of CYCLIN D3;1-10. • Lastly, molecular analysis of soybean germplasms harboring the qSw17-1 QTL for the bigseeded phenotype indicated that reduced expression of GmKIX8-1 is the genetic basis of the qSw17-1 phenotype.

Summary The bacterium Erwinia amylovora, the causal agent of fire blight disease in apple, triggers its infection through the DspA/E effector which interacts with the apple susceptibility protein MdDIPM4. In this work, MdDIPM4 knockout has been produced in two Malus × domestica susceptible cultivars using the CRISPR/Cas9 system delivered via Agrobacterium tumefaciens. Fifty‐seven transgenic lines were screened to identify CRISPR/Cas9‐induced mutations. An editing efficiency of 75% was obtained. Seven edited lines with a loss‐of‐function mutation were inoculated with the pathogen. Highly significant reduction in susceptibility was observed compared to control plants. Sequencing of five potential off‐target sites revealed no mutation event. Moreover, our construct contained a heat‐shock inducible FLP/FRT recombination system designed specifically to remove the T‐DNA harbouring the expression cassettes for CRISPR/Cas9, the marker gene and the FLP itself. Six plant lines with reduced susceptibility to the pathogen were heat‐treated and screened by real‐time PCR to quantify the exogenous DNA elimination. The T‐DNA removal was further validated by sequencing in one plant line. To our knowledge, this work demonstrates for the first time the development and application of a CRISPR/Cas9‐FLP/FRT gene editing system for the production of edited apple plants carrying a minimal trace of exogenous DNA.

Many dominant genetic disorders result from protein-altering mutations, acting primarily through dominant-negative (DN), gain-of-function (GOF), and loss-of-function (LOF) mechanisms. Deciphering the mechanisms by which dominant diseases exert their effects is often experimentally challenging and resource intensive, but is essential for developing appropriate therapeutic approaches. Diseases that arise via a LOF mechanism are more amenable to be treated by conventional gene therapy, whereas DN and GOF mechanisms may require gene editing or targeting by small molecules. Moreover, pathogenic missense mutations that act via DN and GOF mechanisms are more difficult to identify than those that act via LOF using nearly all currently available variant effect predictors. Here, we introduce a tripartite statistical model made up of support vector machine binary classifiers trained to predict whether human protein coding genes are likely to be associated with DN, GOF, or LOF molecular disease mechanisms. We test the utility of the predictions by examining biologically and clinically meaningful properties known to be associated with the mechanisms. Our results strongly support that the models are able to generalise on unseen data and offer insight into the functional attributes of proteins associated with different mechanisms. We hope that our predictions will serve as a springboard for researchers studying novel variants and those of uncertain clinical significance, guiding variant interpretation strategies and experimental characterisation. Predictions for the human UniProt reference proteome are available at https://osf.io/z4dcp/ .

In many plant species, gene dosage is an important cause of phenotype variation. Engineering gene dosage, particularly in polyploid genomes, would provide an efficient tool for plant breeding. The hexaploid oilseed crop Camelina sativa, which has three closely related expressed subgenomes, is an ideal species for investigation of the possibility of creating a large collection of combinatorial mutants. Selective, targeted mutagenesis of the three delta‐12‐desaturase (FAD2) genes was achieved by CRISPR‐Cas9 gene editing, leading to reduced levels of polyunsaturated fatty acids and increased accumulation of oleic acid in the oil. Analysis of mutations over four generations demonstrated the presence of a large variety of heritable mutations in the three isologous CsFAD2 genes. The different combinations of single, double and triple mutants in the T3 generation were isolated, and the complete loss‐of‐function mutants revealed the importance of delta‐12‐desaturation for Camelina development. Combinatorial association of different alleles for the three FAD2 loci provided a large diversity of Camelina lines with various lipid profiles, ranging from 10% to 62% oleic acid accumulation in the oil. The different allelic combinations allowed an unbiased analysis of gene dosage and function in this hexaploid species, but also provided a unique source of genetic variability for plant breeding.