Intelligent fault diagnosis techniques have replaced time-consuming and unreliable human analysis, increasing the efficiency of fault diagnosis. Deep learning models can improve the accuracy of intelligent fault diagnosis with the help of their multilayer nonlinear mapping ability. This paper proposes a novel method named Deep Convolutional Neural Networks with Wide First-layer Kernels (WDCNN). The proposed method uses raw vibration signals as input (data augmentation is used to generate more inputs), and uses the wide kernels in the first convolutional layer for extracting features and suppressing high frequency noise. Small convolutional kernels in the preceding layers are used for multilayer nonlinear mapping. AdaBN is implemented to improve the domain adaptation ability of the model. The proposed model addresses the problem that currently, the accuracy of CNN applied to fault diagnosis is not very high. WDCNN can not only achieve 100% classification accuracy on normal signals, but also outperform the state-of-the-art DNN model which is based on frequency features under different working load and noisy environment conditions.

No abstract

Recent developments in molecular therapies for Duchenne muscular dystrophy (DMD) demand accurate genetic diagnosis, because therapies are mutation specific. The KUCG (Kobe University Clinical Genetics) database for DMD and Becker muscular dystrophy is a hospital-based database comprising 442 cases. Using a combination of complementary DNA (cDNA) and chromosome analysis in addition to conventional genomic DNA-based method, mutation detection was successfully accomplished in all cases, and the largest mutation database of Japanese dystrophinopathy was established. Among 442 cases, deletions and duplications encompassing one or more exons were identified in 270 (61%) and 38 (9%) cases, respectively. Nucleotide changes leading to nonsense mutations or disrupting a splice site were identified in 69 (16%) or 24 (5%) cases, respectively. Small deletion/insertion mutations were identified in 34 (8%) cases. Remarkably, two retrotransposon insertion events were also identified. Dystrophin cDNA analysis successfully revealed novel transcripts with a pseudoexon created by a single-nucleotide change deep within an intron in four cases. X-chromosome abnormalities were identified in two cases. The reading frame rule was upheld for 93% of deletion and 66% of duplication mutation cases. For the application of molecular therapies, induction of exon skipping was deemed the first priority for dystrophinopathy treatment. At one Japanese referral center, the hospital-based mutation database of the dystrophin gene was for the first time established with the highest levels of quality and patient's number.

A few animal studies have shown that wheel running could reverse an unhealthy status by shifting the gut microbial composition, but no investigations have studied the effect of endurance running, such as marathon running, on human gut microbial communities. Since many findings have shown that marathon running immediately causes metabolic changes in blood, urine, muscles and lymph that potentially impact the gut microbiota (GM) within several hours. Here, we investigated whether the GM immediately responds to the enteric changes in amateur half-marathon runners. Alterations in the metabolic profile and microbiota were investigated in fecal samples based on an untargeted metabolomics methodology and 16S rDNA sequencing analysis. A total of 40 fecal metabolites were found significantly changed after finishing a half-marathon race. The most significantly different metabolites were organic acids (the major increased metabolites) and nucleic acid components (the major decreased metabolites). The enteric changes induced by running did not affect the α-diversity of the GM, but the abundances of certain microbiota members were shown to be significantly different before and after running. The family Coriobacteriaceae was identified as a potential biomarker that links exercise with health improvement. Functional prediction showed a significantly activated “Cell motility” function of GM within participants after running. Correlation analysis indicated that the observed differential GM in our study might have been the shared outcome of running and diet. This study provided knowledge regarding the health impacts of marathon running from the perspective of GM for the first time. Our data indicated that long-distance endurance running can immediately cause striking metabolic changes in the gut environment. Gut microbes can rapidly respond to the altered fecal metabolites by adjusting certain bacterial taxa. These findings highlighted the health-promoting benefits of exercise from the perspective of GM.

Myeloid-derived suppressor cells (MDSCs) were one of the major components of the immune suppressive network. STAT3 has an important role in regulating the suppressive potential of MDSCs. In this study, we found that the expression of STAT3 could be modulated by both miR-17-5p and miR-20a. The transfection of miR-17-5p or miR-20a remarkably reduces the expression of reactive oxygen species and the production of H2O2, which are regulated by STAT3. MDSCs transfected with miR-17-5p or miR-20a are less able to suppress Ag-specific CD4 and CD8 T cells. Importantly, both miR-17-5p and miR-20a alleviate the suppressive function of MDSCs in vivo. The expression of miR-17-5p and miR-20a in tumor-associated MDSCs was found to be lower than in Gr1+CD11b+ cells isolated from the spleens of disease-free mice. Tumor-associated factor downregulates the expression of both miR-17-5p and miR-20a. The modulation of miR-17-5p and miR-20a expression may be important for the process by which patients with a tumor can overcome the immune tolerance mediated by MDSCs. Our results suggest that miR-17-5p and miR-20a could potentially be used for immunotherapy against diseases, especially cancer, by blocking STAT3 expression.