Genome-wide analysis of long non-coding RNAs in sugar beet (Beta vulgaris L.) under drought stress.

Drought stress is one of the most severe abiotic stresses that restrict global crop production. Long non-coding RNAs (lncRNAs) have been proved to play a key role in response to drought stress. However, genome-wide identification and characterization of drought-responsive lncRNAs in sugar beet is still lacking. Thus, the present study focused on analyzing lncRNAs in sugar beet under drought stress. We identified 32017 reliable lncRNAs in sugar beet by strand-specific high-throughput sequencing. A total of 386 differentially expressed lncRNAs (DElncRNAs) were found under drought stress. The most significantly upregulated and downregulated lncRNAs were TCONS_00055787 (upregulated by more than 6000 fold) and TCONS_00038334 (downregulated by more than 18000 fold), respectively. Quantitative real-time PCR results exhibited a high concordance with RNA sequencing data, which conformed that the expression patterns of lncRNAs based on RNA sequencing were highly reliable. In addition, we predicted 2353 and 9041 transcripts that were estimated to be the cis- and trans-target genes of the drought-responsive lncRNAs. As revealed by Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, the target genes of DElncRNAs were significantly enriched in organelle subcompartment, thylakoid, endopeptidase activity, catalytic activity, developmental process, lipid metabolic process, RNA polymerase activity, transferase activity, flavonoid biosynthesis and several other terms associated with abiotic stress tolerance. Moreover, 42 DElncRNAs were predicted as potential miRNA target mimics. LncRNAs have important effects on plant adaptation to drought conditions through the interaction with protein-encoding genes. The present study leads to greater insights into lncRNA biology and offers candidate regulators for improving the drought tolerance of sugar beet cultivars at the genetic level.

Intelligent vibration signal denoising method based on non-local fully convolutional neural network for rolling bearings.

Convolutional neural networks (CNNs) have been widely applied to machinery health management in recent years, whereas research on data-driven denoising methods is relatively limited. Therefore, this paper proposes a robust denoising method based on a non-local fully convolutional neural network (NL-FCNN). In this neural network, the Leaky-ReLU activation function is employed to maintain the information contained in the negative value of the signal. The wide kernel principle is also adopted to enlarge the receptive field. Lastly, the non-local means (NLM) is utilized to construct non-local block (NLB), which could efficiently enhance the long-range dependencies learning ability of the network. This block could enormously improve the denoising performance of the network. Moreover, the proposed method exhibits better performance compared with the three conventional denoising methods under multiple noise levels on the Case Western Reserve University (CWRU) motor bearing dataset. Ultimately, we also demonstrate its application to rolling bearing fault diagnosis.