Magnetic separation of peripheral nerve-resident cells underscores key molecular features of human Schwann cells and fibroblasts: an immunochemical and transcriptomics approach

Nerve-derived human Schwann cell (SC) cultures are irreplaceable models for basic and translational research but their use can be limited due to the risk of fibroblast overgrowth. Fibroblasts are an ill-defined population consisting of highly proliferative cells that, contrary to human SCs, do not undergo senescence in culture. We initiated this study by performing an exhaustive immunological and functional characterization of adult nerve-derived human SCs and fibroblasts to reveal their properties and optimize a protocol of magnetic-activated cell sorting (MACS) to separate them effectively both as viable and biologically competent cells. We next used immunofluorescence microscopy imaging, flow cytometry analysis and next generation RNA sequencing (RNA-seq) to unambiguously characterize the post-MACS cell products. High resolution transcriptome profiling revealed the identity of key lineage-specific transcripts and the clearly distinct neural crest and mesenchymal origin of human SCs and fibroblasts, respectively. Our analysis underscored a progenitor- or stem cell-like molecular phenotype in SCs and fibroblasts and the heterogeneity of the fibroblast populations. In addition, pathway analysis of RNA-seq data highlighted putative bidirectional networks of fibroblast-to-SC signaling that predict a complementary, yet seemingly independent contribution of SCs and fibroblasts to nerve regeneration. In sum, combining MACS with immunochemical and transcriptomics approaches provides an ideal workflow to exhaustively assess the identity, the stage of differentiation and functional features of highly purified cells from human peripheral nerve tissues.

End-to-end recognition of streaming Japanese speech using CTC and local attention

Many end-to-end, large vocabulary, continuous speech recognition systems are now able to achieve better speech recognition performance than conventional systems. Most of these approaches are based on bidirectional networks and sequence-to-sequence modeling however, so automatic speech recognition (ASR) systems using such techniques need to wait for an entire segment of voice input to be entered before they can begin processing the data, resulting in a lengthy time-lag, which can be a serious drawback in some applications. An obvious solution to this problem is to develop a speech recognition algorithm capable of processing streaming data. Therefore, in this paper we explore the possibility of a streaming, online, ASR system for Japanese using a model based on unidirectional LSTMs trained using connectionist temporal classification (CTC) criteria, with local attention. Such an approach has not been well investigated for use with Japanese, as most Japanese-language ASR systems employ bidirectional networks. The best result for our proposed system during experimental evaluation was a character error rate of 9.87%.

Comparación entre redes de fibras sintéticas y redes de fibras de cáñamo para el refuerzo de muros de albañileria = Comparison between synthetic fiber networks and hemp fiber networks for the reinforcement of masonry walls

ResumenLa existencia de una gran cantidad de edificios antiguos ha movilizado la investigación para estudiar nuevos sistemas de refuerzo a aquellas construcciones que sean dañadas por decadencia fisiológica o por terremoto. El presente documento demuestra la eficacia de un sistema de refuerzo innovador para los muros de mampostería, constituyentes los elementos estructurales de los edificios históricos. Se hace una comparación entre unas redes bidireccionales de fibras sintéticas tradicionales y las de fibras naturales, ambas pegadas a las dos fachadas del muro con matriz de mortero. El resultado muestra la aplicación de las fibras de cáñamo como refuerzo sísmico y una mayor compatibilidad de estas con el material que caracterizan los edificios antiguos. AbstractThe existence of a large number of old buildings has mobilized research to study new systems of reinforcement to those buildings that are damaged by physiological decay or earthquake. This document demonstrates the effectiveness of an innovative reinforcement system for masonry walls, which are the structural elements of historic buildings. A comparison is made between bidirectional networks of traditional synthetic fibers and those of natural fibers, both glued to the two facades of the wall with mortar matrix. The result shows the application of hemp fibers as seismic reinforcement and a greater compatibility of these with the material that characterize the old buildings.