We present a new nonlinear mode decomposition method to visualize the decomposed flow fields, named the mode decomposing convolutional neural network (MD-CNN). The proposed method is applied to a flow around a circular cylinder at Re D = 100 as a test case. The flow attributes are mapped into two modes in the latent space and then these two modes are visualized in the physical space. Since the MD-CNNs with nonlinear activation functions show lower reconstruction errors than the proper orthogonal decomposition (POD), the nonlinearity contained in the activation function is considered the key to improve the capability of the model. It is found by applying POD to each field decomposed using the MD-CNN with hyperbolic tangent activation that a single nonlinear MD-CNN mode contains multiple orthogonal bases, in contrast to the linear methods, i.e., POD and MD-CNN with linear activation. The present results suggest a great potential for the nonlinear MD-CNN to be used for feature extraction of flow fields in lower dimension than POD, while retaining interpretable relationships with the conventional POD modes.

The development of axons and dendrites is controlled by small GTP-binding proteins of the Rho family, but the upstream signaling mechanisms responsible for such regulation remain unclear. We have now investigated the role of the transmembrane protein cluster of differentiation 47 (CD47) in this process with hippocampal neurons. CD47-deficient neurons manifested markedly impaired development of dendrites and axons, whereas overexpression of CD47 promoted such development. Interaction of SH2 domain-containing protein tyrosine phosphatase substrate-1 (SHPS-1) with CD47 also induced the formation of dendritic filopodia and spines. These effects of CD47 were prevented by inhibition of either cell division cycle 42 (Cdc42) or Rac. In CD47-deficient neurons, autophosphorylation of Src was markedly reduced. In addition, overexpression of CD47 promoted the autophosphorylation of Src. Inhibition of Src family kinases indeed prevented CD47-promoted dendritic development. Inhibition of either FGD1-related Cdc42-guanine nucleotide exchange factor (GEF) (FRG) or Vav2, which is a GEF for Cdc42 and Rac and is activated by Src, also prevented the effects of CD47 on dendritic development. These results indicate that CD47 promotes development of dendrites and axons in hippocampal neurons in a manner dependent, at least in part, on activation of Cdc42 and Rac mediated by Src as well as by FRG and Vav2.

No abstract

The present large-scale clinical survey revealed current epidemiological trends for idiopathic sudden sensorineural hearing loss (SSNHL) and various factors associated with the severity of hearing impairment and prognosis.

The vancomycin-resistant enterococci GV1, GV2 and GV3, which were isolated from droppings from broiler farms in Japan have been characterized as VanA-type VRE, which express high-level vancomycin resistance (256 or 512 microg ml(-1), MIC) and low-level teicoplanin resistance (1 or 2 microg ml(-1), MIC). The vancomycin resistances were encoded on plasmids. The vancomycin resistance conjugative plasmid pMG2 was isolated from the GV2 strain. The VanA determinant of pMG2 showed the same genetic organization as that of the VanA genes encoded on the representative transposon Tn1546, which comprises vanRSHAXYZ. The nucleotide sequences of all the genes, except the gene related to the vanS gene on Tn1546, were completely identical to the genes encoded on Tn1546. Three amino acid substitutions in the N-terminal region of the deduced VanS were detected in the nucleotide sequence of vanS encoded on pMG2. There were also three amino acid substitutions in the vanS gene of the GV1 and GV3 strains in the same positions as in the vanS gene of pMG2. Vancomycin induced the increased teicoplanin resistance in these strains.

We investigate the capability of machine learning (ML) based reduced order model (ML-ROM) for two-dimensional unsteady flows around a circular cylinder at different Reynolds numbers. The present ML-ROM is constructed by two ML schemes: a convolutional neural network-based autoencoder (CNN-AE) and a long short-term memory (LSTM). The CNN-AE is utilized to map high-dimensional flow fields obtained by direct numerical simulation (DNS) into a low-dimensional latent space while keeping their spatially coherent information. The LSTM is then trained to learn the temporal evolution of the mapped latent vectors together with the information on the Reynolds number. Using the trained LSTM model, the high-dimensional dynamics of flow fields can be reproduced with the aid of the decoder part of CNN-AE, which can map the predicted low-dimensional latent vector to the high-dimensional space. We find that the flow fields generated by the present ML-ROM show statistical agreement with the reference DNS data. The dependence of the accuracy of the proposed model on the Reynolds number is also examined in detail. The present results demonstrate that the ML-ROM can reconstruct flows at the Reynolds numbers that were not used in the training process unless the flow regime drastically changes.

A variant in a transcription factor gene, POU4F3, is responsible for autosomal dominant nonsyndromic hereditary hearing loss, DFNA15. To date, 14 variants, including a whole deletion of POU4F3, have been reported to cause HL in various ethnic groups. In the present study, genetic screening for POU4F3 variants was carried out for a large series of Japanese hearing loss (HL) patients to clarify the prevalence and clinical characteristics of DFNA15 in the Japanese population. Massively parallel DNA sequencing of 68 target candidate genes was utilized in 2,549 unrelated Japanese HL patients (probands) to identify genomic variations responsible for HL. The detailed clinical features in patients with POU4F3 variants were collected from medical charts and analyzed. Novel 12 POU4F3 likely pathogenic variants (six missense variants, three frameshift variants, and three nonsense variants) were successfully identified in 15 probands (2.5%) among 602 families exhibiting autosomal dominant HL, whereas no variants were detected in the other 1,947 probands with autosomal recessive or inheritance pattern unknown HL. To obtain the audiovestibular configuration of the patients harboring POU4F3 variants, we collected audiograms and vestibular symptoms of the probands and their affected family members. Audiovestibular phenotypes in a total of 24 individuals from the 15 families possessing variants were characterized by progressive HL, with a large variation in the onset age and severity with or without vestibular symptoms observed. Pure-tone audiograms indicated the most prevalent configuration as mid-frequency HL type followed by high-frequency HL type, with asymmetry observed in approximately 20% of affected individuals. Analysis of the relationship between age and pure-tone average suggested that individuals with truncating variants showed earlier onset and slower progression of HL than did those with non-truncating variants. The present study showed that variants in POU4F3 were a common cause of autosomal dominant HL.

Tumor-associated macrophages (TAM) have been classified into an immunostimulatory M1 subset against microbes and malignancies, and an immunoregulatory M2 subset that secretes immunosuppressive cytokines in order to repair tissues damaged by malignancies. The infiltration of M2 in the tumor microenvironment is known to facilitate immunosuppression and tumor-promoting properties. In the present study, we investigated the phagocytic potential of these macrophage subsets in oral squamous cell carcinoma (OSCC) in relation to the expression of CD47, the 'don't eat me' signal against macrophages. The macrophage subsets M1 (induced by GM-CSF and IFN-γ) and M2 (induced by M-CSF and IL-10) were derived from the CD14(+) cells of healthy donors. Phagocytosis of the CFSE-labeled CD47(+) cell line HSC-3 by M1/M2 was assessed using flow cytometry and suppressed by an anti-CD47 neutralizing antibody or CD47 siRNA. Furthermore, CD68(+) and CD163(+) macrophage subset counts infiltrating tumor tissue and the expression of CD47 on cancer cells were examined immunohistochemically in 74 cases of OSCC, and their relationships with clinicopathological parameters or prognoses were determined. The phagocytic potential of M1 was similar to that of M2 in vitro. Phagocytosis by M1 increased in a CD47-dependent manner by the neutralizing antibody and siRNA, but did not in M2. An immunohistochemical (IHC) analysis revealed that the expression of CD47 did not correlate with macrophage subsets in peritumoral tissue or with any clinicopathological parameters; however, the stronger expression of CD47 by cancer cells and larger number of total macrophages/M2 were independently related to shorter survivals. Our results suggest that the expression of CD47 by cancer cells is related to evasion from phagocytosis, particularly that by M1 in vitro. IHC results indicate that various mechanisms are involved in the engulfing potential of TAM subsets in vivo.