Explore the vast range of titles available.

In this study, bearing fault diagnosis is performed with a small amount of data through few-shot learning. Recently, a fault diagnosis method based on deep learning has achieved promising results. Most studies required numerous training samples for fault diagnosis. However, at manufacturing sites, it is impossible to have enough training samples to represent all fault types under all operating conditions. In addition, most studies consider only accuracy, and models are complex and computationally expensive. Research that only considers accuracy is inefficient since manufacturing sites change rapidly. Therefore, in this study, we propose a few-shot learning model that can effectively learn with small data. In addition, a Depthwise Separable Convolution layer that can effectively reduce parameters is used together. In order to find an efficient model, the optimal hyperparameters were found by adjusting the number of blocks and hyperparameters, and by using a Depthwise Separable Convolution layer for the optimal hyperparameters, it showed higher accuracy and fewer parameters than the existing model.

Background Microorganisms are important occupants of many different environments. Identifying the composition of microbes and estimating their abundance promote understanding of interactions of microbes in environmental samples. To understand their environments more deeply, the composition of microorganisms in environmental samples has been studied using metagenomes, which are the collections of genomes of the microorganisms. Although many tools have been developed for taxonomy analysis based on different algorithms, variability of analysis outputs of existing tools from the same input metagenome datasets is the main obstacle for many researchers in this field. Results Here, we present a novel meta-analysis tool for metagenome taxonomy analysis, called TAMA, by intelligently integrating outputs from three different taxonomy analysis tools. Using an integrated reference database, TAMA performs taxonomy assignment for input metagenome reads based on a meta-score by integrating scores of taxonomy assignment from different taxonomy classification tools. TAMA outperformed existing tools when evaluated using various benchmark datasets. It was also successfully applied to obtain relative species abundance profiles and difference in composition of microorganisms in two types of cheese metagenome and human gut metagenome. Conclusion TAMA can be easily installed and used for metagenome read classification and the prediction of relative species abundance from multiple numbers and types of metagenome read samples. TAMA can be used to more accurately uncover the composition of microorganisms in metagenome samples collected from various environments, especially when the use of a single taxonomy analysis tool is unreliable. TAMA is an open source tool, and can be downloaded at https://github.com/jkimlab/TAMA .

Infectious diseases caused by diverse pathogens—including viruses, bacteria, and fungi—pose a significant diagnostic challenge, especially when conventional assays fail to identify the causative agents. Next-generation sequencing (NGS), particularly hybrid-capture-based NGS, offers a culture-independent approach for comprehensive pathogen detection directly from clinical samples. In this study, we developed a hybrid-capture-based NGS panel targeting 85 respiratory pathogens (62 viruses, 23 bacteria/fungi) and evaluated its analytical and clinical performance. Analytical validation using 50 pathogens demonstrated high sensitivity, with 10× genome coverage ≥ 96% in 32 of 36 viral targets and ≥ 95% in all 14 bacterial/fungal targets. The panel maintained ≥ 80% genome coverage at Ct values up to 34. Clinical assessment of 561 respiratory specimens identified 25 pathogen species, achieving ≥ 90% genome coverage in 57.5% of virus-positive and 83.6% of bacteria-positive samples. Notably, additional pathogens were detected in high Ct value undetected by RT-PCR, and 35 cases of viral–viral and viral–bacterial co-infections were identified. This approach improves the detection performance for respiratory infections of unknown etiology and serves as a practical platform for the detection of emerging pathogens and public health preparedness. This technology has significant potential for applications in infectious disease surveillance and development of vaccines and therapeutics.

Background Many studies have been performed to identify various genomic loci and genes associated with the meat quality in pigs. However, the full genetic architecture of the trait still remains unclear in part because of the lack of accurate identification of related structural variations (SVs) which resulted from the shortage of target breeds, the limitations of sequencing data, and the incompleteness of genome assemblies. The recent generation of a new pig breed with superior meat quality, called Nanchukmacdon, and its chromosome-level genome assembly (the NCMD assembly) has provided new opportunities. Results By applying assembly-based SV calling approaches to various genome assemblies of pigs including Nanchukmacdon, the impact of SVs on meat quality was investigated. Especially, by checking the commonality of SVs with other pig breeds, a total of 13,819 Nanchukmacdon-specific SVs (NSVs) were identified, which have a potential effect on the unique meat quality of Nanchukmacdon. The regulatory potentials of NSVs for the expression of nearby genes were further examined using transcriptome- and epigenome-based analyses in different tissues. Conclusions Whole-genome comparisons based on chromosome-level genome assemblies have led to the discovery of SVs affecting meat quality in pigs, and their regulatory potentials were analyzed. The identified NSVs will provide new insights regarding genetic architectures underlying the meat quality in pigs. Finally, this study confirms the utility of chromosome-level genome assemblies and multi-omics analysis to enhance the understanding of unique phenotypes.

This study aimed to develop and validate a transformer-based early warning score (TEWS) system for predicting adverse events (AEs) in the emergency department (ED). We conducted a retrospective study analyzing adult ED visits at a tertiary hospital. The TEWS was developed to predict five AEs within 24 h: vasopressor use, respiratory support, intensive care unit admission, septic shock, and cardiac arrest. Performance was evaluated and compared using the area under the receiver operating characteristic curve (AUROC) and bootstrap-based t-test. External validation was performed using the Marketplace for Medical Information in Intensive Care (MIMIC)-IV-ED database. Transfer learning was applied using 1% and 5% of the external data. A total of 414,748 patients was analyzed in the development cohort (AEs, 3.7%), and 410,880 patients (AEs, 6.7%) were included in the external validation cohort. Compared to the modified early warning score (MEWS), the TEWS incorporating 13 variables and the vital signs-only TEWS demonstrated superior prognostic performance across all AEs. The AUROC ranged from 0.833 to 0.936 for TEWS and 0.688 to 0.874 for MEWS. In external validation, the TEWS also showed acceptable discrimination with AUROC values of 0.759 to 0.905. Transfer learning significantly improved the performance, increasing AUROC values to 0.846–0.911. The TEWS system was successfully integrated into the electronic health record (EHR) system of the study hospital, providing real-time risk assessment for ED patients. We developed and validated an artificial intelligence-based early warning score system that predicts multiple adverse outcomes in the ED and was successfully integrated into the EHR system.

Background Advances in next-generation sequencing technologies have provided an opportunity to perform population-level comparative genomic analysis to discover unique genomic characteristics of domesticated animals. Duck is one of the most popular domesticated waterfowls, which is economically important as a source of meat, eggs, and feathers. The objective of this study is to perform population and functional analyses of Korean native duck, which has a distinct meat flavor and texture phenotype, using whole-genome sequencing data. To study the distinct genomic features of Korean native duck, we conducted population-level genomic analysis of 20 Korean native ducks together with 15 other duck breeds. Results A total of 15.56 million single nucleotide polymorphisms were detected in Korean native duck. Based on the unique existence of non-synonymous single nucleotide polymorphisms in Korean native duck, a total of 103 genes related to the unique genomic characteristics of Korean native duck were identified in comparison with 15 other duck breeds, and their functions were investigated. The nucleotide diversity and population structures among the used duck breeds were then compared, and their phylogenetic relationship was analyzed. Finally, highly differentiated genomic regions among Korean native duck and other duck breeds were identified, and functions of genes in those regions were examined. Conclusions This is the first study to compare the population of Korean native duck with those of other duck breeds by using whole-genome sequencing data. Our findings can be used to expand our knowledge of genomic characteristics of Korean native duck, and broaden our understanding of duck breeds.

Background Advances in sequencing technologies have facilitated large-scale comparative genomics based on whole genome sequencing. Constructing and investigating conserved genomic regions among multiple species (called synteny blocks) are essential in the comparative genomics. However, they require significant amounts of computational resources and time in addition to bioinformatics skills. Many web interfaces have been developed to make such tasks easier. However, these web interfaces cannot be customized for users who want to use their own set of genome sequences or definition of synteny blocks. Results To resolve this limitation, we present mySyntenyPortal, a stand-alone application package to construct websites for synteny block analyses by using users’ own genome data. mySyntenyPortal provides both command line and web-based interfaces to build and manage websites for large-scale comparative genomic analyses. The websites can be also easily published and accessed by other users. To demonstrate the usability of mySyntenyPortal, we present an example study for building websites to compare genomes of three mammalian species (human, mouse, and cow) and show how they can be easily utilized to identify potential genes affected by genome rearrangements. Conclusions mySyntenyPortal will contribute for extended comparative genomic analyses based on large-scale whole genome sequences by providing unique functionality to support the easy creation of interactive websites for synteny block analyses from user’s own genome data.

Pig as a food source serves daily dietary demand to a wide population around the world. Preference of meat depends on various factors with muscle play the central role. In this regards, selective breeding abled us to develop “Nanchukmacdon” a pig breeds with an enhanced variety of meat and high fertility rate. To identify genomic regions under selection we performed whole-genome resequencing, transcriptome, and whole-genome bisulfite sequencing from Nanchukmacdon muscles samples and used published data for three other breeds such as Landrace, Duroc, Jeju native pig and analyzed the functional characterization of candidate genes. In this study, we present a comprehensive approach to identify candidate genes by using multi-omics approaches. We performed two different methods XP-EHH, XP-CLR to identify traces of artificial selection for traits of economic importance. Moreover, RNAseq analysis was done to identify differentially expressed genes in the crossed breed population. Several genes ( UGT8, ZGRF1, NDUFA10, EBF3, ELN, UBE2L6, NCALD, MELK, SERP2, GDPD5, and FHL2 ) were identified as selective sweep and differentially expressed in muscles related pathways. Furthermore, nucleotide diversity analysis revealed low genetic diversity in Nanchukmacdon for identified genes in comparison to related breeds and whole-genome bisulfite sequencing data shows the critical role of DNA methylation pattern in identified genes that leads to enhanced variety of meat. This work demonstrates a way to identify the molecular signature and lays a foundation for future genomic enabled pig breeding.

Genomic data have become major resources to understand complex mechanisms at fine-scale temporal and spatial resolution in functional and evolutionary genetic studies, including human diseases, such as cancers. Recently, a large number of whole genomes of evolving populations of yeast (Saccharomyces cerevisiae W303 strain) were sequenced in a time-dependent manner to identify temporal evolutionary patterns. For this type of study, a chromosome-level sequence assembly of the strain or population at time zero is required to compare with the genomes derived later. However, there is no fully automated computational approach in experimental evolution studies to establish the chromosome-level genome assembly using unique features of sequencing data. In this study, we developed a new software pipeline, the integrative meta-assembly pipeline (IMAP), to build chromosome-level genome sequence assemblies by generating and combining multiple initial assemblies using three de novo assemblers from short-read sequencing data. We significantly improved the continuity and accuracy of the genome assembly using a large collection of sequencing data and hybrid assembly approaches. We validated our pipeline by generating chromosome-level assemblies of yeast strains W303 and SK1, and compared our results with assemblies built using long-read sequencing and various assembly evaluation metrics. We also constructed chromosome-level sequence assemblies of S. cerevisiae strain Sigma1278b, and three commonly used fungal strains: Aspergillus nidulans A713, Neurospora crassa 73, and Thielavia terrestris CBS 492.74, for which long-read sequencing data are not yet available. Finally, we examined the effect of IMAP parameters, such as reference and resolution, on the quality of the final assembly of the yeast strains W303 and SK1. We developed a cost-effective pipeline to generate chromosome-level sequence assemblies using only short-read sequencing data. Our pipeline combines the strengths of reference-guided and meta-assembly approaches. Our pipeline is available online at http://github.com/jkimlab/IMAP including a Docker image, as well as a Perl script, to help users install the IMAP package, including several prerequisite programs. Users can use IMAP to easily build the chromosome-level assembly for the genome of their interest.

Chronic kidney disease (CKD) is a prevalent condition with significant global health implications. Early detection and management are critical to prevent disease progression and complications. Deep learning (DL) models using retinal images have emerged as potential noninvasive screening tools for CKD, though their performance may be limited, especially in identifying individuals with proteinuria and in specific subgroups. We aim to evaluate the efficacy of integrating retinal images and urine dipstick data into DL models for enhanced CKD diagnosis. The 3 models were developed and validated: eGFR-RIDL (estimated glomerular filtration rate-retinal image deep learning), eGFR-UDLR (logistic regression using urine dipstick data), and eGFR-MMDL (multimodal deep learning combining retinal images and urine dipstick data). All models were trained to predict an eGFR<60 mL/min/1.73 m², a key indicator of CKD, calculated using the 2009 CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) equation. This study used a multicenter dataset of participants aged 20-79 years, including a development set (65,082 people) and an external validation set (58,284 people). Wide Residual Networks were used for DL, and saliency maps were used to visualize model attention. Sensitivity analyses assessed the impact of numerical variables. eGFR-MMDL outperformed eGFR-RIDL in both the test and external validation sets, with area under the curves of 0.94 versus 0.90 and 0.88 versus 0.77 (P<.001 for both, DeLong test). eGFR-UDLR outperformed eGFR-RIDL and was comparable to eGFR-MMDL, particularly in the external validation. However, in the subgroup analysis, eGFR-MMDL showed improvement across all subgroups, while eGFR-UDLR demonstrated no such gains. This suggested that the enhanced performance of eGFR-MMDL was not due to urine data alone, but rather from the synergistic integration of both retinal images and urine data. The eGFR-MMDL model demonstrated the best performance in individuals younger than 65 years or those with proteinuria. Age and proteinuria were identified as critical factors influencing model performance. Saliency maps indicated that urine data and retinal images provide complementary information, with urine offering insights into retinal abnormalities and retinal images, particularly the arcade vessels, being key for predicting kidney function. The MMDL model integrating retinal images and urine dipstick data show significant promise for noninvasive CKD screening, outperforming the retinal image-only model. However, routine blood tests are still recommended for individuals aged 65 years and older due to the model's limited performance in this age group.