Yu-Wei Wu

In current genomic research, the widely used methods for predicting antimicrobial resistance (AMR) often rely on prior knowledge of known AMR genes or reference genomes. However, these methods have limitations, potentially resulting in imprecise predictions owing to incomplete coverage of AMR mechanisms and genetic variations. To overcome these limitations, we propose a pan-genome-based machine learning approach to advance our understanding of AMR gene repertoires and uncover possible feature sets for precise AMR classification. By building compacted de Brujin graphs (cDBGs) from thousands of genomes and collecting the presence/absence patterns of unique sequences (unitigs) for Pseudomonas aeruginosa, we determined that using machine learning models on unitig-centered pan-genomes showed significant promise for accurately predicting the antibiotic resistance or susceptibility of …

This study aimed to develop and externally validate a prognostic prediction model for screening fetal growth restriction (FGR)/small for gestational age (SGA) using medical history. From a nationwide health insurance database (n= 1,697,452), we retrospectively selected visits of 12-to-55-year-old females to healthcare providers. This study used machine learning (including deep learning) and 54 medical-history predictors. The best model was a deep-insight visible neural network (DI-VNN). It had area under the curve of receiver operating characteristics (AUROC) 0.742 (95% CI 0.734 to 0.750) and a sensitivity of 49.09%(95% CI 47.60% to 50.58% at with 95% specificity). Our model used medical history for screening FGR/SGA with moderate accuracy by DI-VNN. In future work, we will compare this model with those from systematically-reviewed, previous studies and evaluate if this model's usage impacts patient …

Objectives Prevention of fetal growth restriction/small for gestational age is adequate if screening is accurate. Ultrasound and biomarkers can achieve this goal; however, both are often inaccessible. This study aimed to develop, validate, and deploy a prognostic prediction model for screening fetal growth restriction/small for gestational age using only medical history. Methods From a nationwide health insurance database (n=1,697,452), we retrospectively selected visits of 12-to-55-year-old females to 22,024 healthcare providers of primary, secondary, and tertiary care. This study used machine learning (including deep learning) to develop prediction models using 54 medical-history predictors. After evaluating model calibration, clinical utility, and explainability, we selected the best by discrimination ability. We also externally validated and compared the models with those from previous studies, which were rigorously selected by a systematic review of Pubmed, Scopus, and Web of Science. Results We selected 169,746 subjects with 507,319 visits for predictive modeling. The best prediction model was a deep-insight visible neural network. It had an area under the receiver operating characteristics curve of 0.742 (95% confidence interval 0.734 to 0.750) and a sensitivity of 49.09% (95% confidence interval 47.60% to 50.58% using a threshold with 95% specificity). The model was competitive against the previous models in a systematic review of 30 eligible studies of 381 records, including those using either ultrasound or biomarker measurements. We deployed a web application to apply the model. Conclusions Our model used only medical history …

In contemporary biomedical research, the accurate automatic detection of cells within intricate microscopic imagery stands as a cornerstone for scientific advancement. Leveraging state-of-the-art deep learning techniques, this study introduces a novel amalgamation of Fuzzy Automatic Contrast Enhancement (FACE) and the You Only Look Once (YOLO) framework to address this critical challenge of automatic cell detection. Yeast cells, representing a vital component of the fungi family, hold profound significance in elucidating the intricacies of eukaryotic cells and human biology. The proposed methodology introduces a paradigm shift in cell detection by optimizing image contrast through optimal fuzzy clustering within the FACE approach. This advancement mitigates the shortcomings of conventional contrast enhancement techniques, minimizing artifacts and suboptimal outcomes. Further enhancing contrast, a …

Biofuels represent a promising path toward a future less reliant on fossil fuels. One reason why is because the components to generate biofuels can be found in ordinary waste, such as compost, where degradative microbes unleash the energy-dense sugars found in plant matter. While scouting for the hungriest degraders marks one way of boosting biofuel production, understanding how microbes work together to devour plant matter could make the greatest impact. In a recent study, researchers examined three microbial communities derived from the same green-waste compost. Despite their shared origins, in two weeks, each had evolved a unique sorghum-degrading system. For example, while all three communities were rich in bacteria of the genus Actinotalea, Community 2 possessed more Actinotalea-related degradative enzymes. Yet Communities 1 and 3 showed higher performance in terms of biomass …

Background: Predicting the resistance profiles of antimicrobial resistance (AMR) pathogens is becoming more and more important in treating infectious diseases. Various attempts have been made to build machine learning models to classify resistant or susceptible pathogens based on either known antimicrobial resistance genes or the entire gene set. However, the phenotypic annotations are translated from minimum inhibitory concentration (MIC), which is the lowest concentration of antibiotic drugs in inhibiting certain pathogenic strains. Since the MIC breakpoints that classify a strain to be resistant or susceptible to specific antibiotic drug may be revised by governing institutes, we refrained from translating these MIC values into the categories “susceptible” or “resistant” but instead attempted to predict the MIC values using machine learning approaches. Results: By applying a machine learning feature selection approach on a Salmonella enterica pan-genome, in which the protein sequences were clustered to identify highly similar gene families, we showed that the selected features (genes) performed better than known AMR genes, and that models built on the selected genes achieved very accurate MIC prediction. Functional analysis revealed that about half of the selected genes were annotated as hypothetical proteins (i.e., with unknown functional roles), and that only a small portion of known AMR genes were among the selected genes, indicating that applying feature selection on the entire gene set has the potential of uncovering novel genes that may be associated with and may contribute to pathogenic antimicrobial resistances …

We aimed to provide a resampling protocol for dimensional reduction resulting a few latent variables. The applicability focuses on but not limited for developing a machine learning prediction model in order to improve the number of sample size in relative to the number of candidate predictors. By this feature representation technique, one can improve generalization by preventing latent variables to overfit data used to conduct the dimensional reduction. However, this technique may warrant more computational capacity and time to conduct the procedure. The key stages consisted of derivation of latent variables from multiple resampling subsets, parameter estimation of latent variables in population, and selection of latent variables transformed by the estimated parameters.

Abstract Background and Objective: Deep learning is applied in medicine mostly due to its state-of-the-art performance for diagnostic imaging. Supervisory authorities also require the model to be explainable, but most explain the model after development (post hoc) instead of incorporating explanation into the design (ante hoc). This study aimed to demonstrate a human-guided deep learning with ante-hoc explainability by convolutional network from non-image data to develop, validate, and deploy a prognostic prediction model for PROM and an estimator of time of delivery using a nationwide health insurance database. Methods: To guide modeling, we constructed and verified association diagrams respectively from literatures and electronic health records. Non-image data were transformed into meaningful images utilizing predictor-to-predictor similarities, harnessing the power of convolutional neural network …

Abscondita cerata is the most abundant and widely distributed endemic firefly species in Taiwan and is considered a key environmental and ecological indicator organism. In this study, we report the first long-read genome sequencing of Abs. cerata sequenced by Nanopore technology. The draft genome size, 967 Mb, was measured through a hybrid approach that consisted of assembling using 11.25-Gb Nanopore long reads and polishing using 9.47-Gb BGI PE100 short reads. The drafted genome was assembled into 4,855 contigs, with the N50 reaching 325.269 kb length. The assembled genome was predicted to possess 55,206 protein-coding genes, of which 20,862 (37.78%) were functionally annotated with public databases. 47.11% of the genome sequences consisted of repeat elements; among them DNA transposons accounted for the largest proportion (26.79%). A BUSCO (Benchmarking Universal …

The traditional natural product discovery approach has accessed only a fraction of the chemical diversity in nature. The use of bioinformatic tools to interpret the instructions encoded in microbial biosynthetic genes has the potential to circumvent the existing methodological bottlenecks and greatly expand the scope of discovery. Structural prediction algorithms for nonribosomal peptides (NRPs), the largest family of microbial natural products, lie at the heart of this new approach. To understand the scope and limitation of the existing prediction algorithms, we evaluated their performances on NRP synthetase biosynthetic gene clusters. Our systematic analysis shows that the NRP biosynthetic landscape is uneven. Phenylglycine and its derivatives as a group of NRP building blocks (BBs), for example, have been oversampled, reflecting an extensive historical interest in the glycopeptide antibiotics family. In contrast, the …

The partial nitritation-anaerobic ammonium oxidation (anammox; PN-A) process has been considered a sustainable method for wastewater ammonium removal, with recent attempts to treat low-strength wastewater. However, how microbes adapt to the alternate microaerobic-anoxic operation of the process when treating low ammonium concentrations remains poorly understood. In this study, we applied a metagenomic approach to determine the genomic contents of core members in a PN-A reactor treating inorganic ammonium wastewater at loading as low as 0.0192 kg-N/m3/day. The metabolic traits of metagenome-assembled genomes from 18 core species were analyzed. Taxonomically diverse ammonia oxidizers, including two Nitrosomonas species, a comammox Nitrospira species, a novel Chloroflexota-related species, and two anammox bacteria, Ca. Brocadia and Ca. Jettenia, accounted for the PN-A reactions. The characteristics of a series of genes encoding class II ribonucleotide reductase, high-affinity bd-type terminal oxidase, and diverse antioxidant enzymes revealed that comammox Nitrospira has a superior adaptation ability over the competitors, which may confer the privileged partnership with anammox bacteria in the PN-A reactor. This finding is supported by the long-term monitoring experiment, showing the predominance of the comammox Nitrospira in the ammonia-oxidizing community. Metagenomic analysis of seven heterotrophs suggested that nitrate reduction is a common capability in potentially using endogenous carbohydrates and peptides to enhance nitrogen removals. The prevalence of class II ribonucleotide …

Background Existing proposed pathogenesis for preeclampsia (PE) was only applied for early onset subtype and did not consider pre-pregnancy and competing risks. We aimed to decipher PE subtypes by identifying related transcriptome that represents endometrial maturation and histologic chorioamnionitis. Methods We utilized eight arrays of mRNA expression for discovery (n=289), and other eight arrays for validation (n=352). Differentially expressed genes (DEGs) were overlapped between those of: (1) healthy samples from endometrium, decidua, and placenta, and placenta samples under histologic chorioamnionitis; and (2) placenta samples for each of the subtypes. They were all possible combinations based on four axes: (1) pregnancy-induced hypertension; (2) placental dysfunction-related diseases (e.g., fetal growth restriction [FGR]); (3) onset; and (4) severity. Results The DEGs of endometrium at late …

Understanding genes and their underlying mechanisms is critical in deciphering how antimicrobial-resistant (AMR) bacteria withstand detrimental effects of antibiotic drugs. At the same time the genes related to AMR phenotypes may also serve as biomarkers for predicting whether a microbial strain is resistant to certain antibiotic drugs. We developed a Cross-Validated Feature Selection (CVFS) approach for robustly selecting the most parsimonious gene sets for predicting AMR activities from bacterial pan-genomes. The core idea behind the CVFS approach is interrogating features among non-overlapping sub-parts of the datasets to ensure the representativeness of the features. By randomly splitting the dataset into disjoint sub-parts, conducting feature selection within each sub-part, and intersecting the features shared by all sub-parts, the CVFS approach is able to achieve the goal of extracting the most …

Gut microbial proteolytic metabolism has been reportedly altered in Parkinson’s disease (PD). However, the circulating aromatic amino acids (AAA) described in PD are inconsistent. Here we aimed to investigate plasma AAA profiles in a large cohort of PD patients, and examine their correlations with clinical severity and gut microbiota changes. We enrolled 500 participants including 250 PD patients and 250 neurologically normal controls. Plasma metabolites were measured using liquid chromatography mass spectrometry. Faecal samples were newly collected from 154 PD patients for microbiota shotgun metagenomic sequencing combined with data derived from 96 PD patients reported before. Data were collected regarding diet, medications, and motor and non-motor symptoms of PD. Compared to controls, PD patients had higher plasma AAA levels, including phenylacetylglutamine (PAGln), p-cresol sulfate …

Recently, human activity recognition (HAR) techniques have made remarkable developments in the field of machine learning. In this paper, we classify human gestures using data collected from a curved piezoelectric sensor, including elbow movement, wrist turning, wrist bending, coughing, and neck bending. The classification process relies on data collected from a sensor. Machine learning algorithms enabled with K-mer are developed and optimized to perform human gesture recognition (HGR) from the acquired data to achieve the best results. Three machine learning algorithms, namely support vector machine (SVM), random forest (RF), and k-nearest neighbor (k-NN), are performed and analyzed with K-mer. The input parameters such as subsequence length (K), number of cuts, penalty parameter (C), number of trees (n_estimators), maximum depth of the tree (max_depth), and nearest neighbors (k) for the three machine learning algorithms are modified and analyzed for classification accuracy. The proposed model was evaluated using its accuracy percentage, recall score, precision score, and F-score value. We achieve promising results with accuracy of 94.11 ± 0.3%, 97.18 ± 0.4%, and 96.90 ± 0.5% for SVM, RF, and k-NN, respectively. The execution time to run the program with optimal parameters is 19.395 ± 1 s, 5.941 ± 1 s, and 3.832 ± 1 s for SVM, RF, and k-NN, respectively.

BackgroundPlant cell walls are interwoven structures recalcitrant to degradation. Native and adapted microbiomes can be particularly effective at plant cell wall deconstruction. Although most understanding of biological cell wall deconstruction has been obtained from isolates, cultivated microbiomes that break down cell walls have emerged as new sources for biotechnologically relevant microbes and enzymes. These microbiomes provide a unique resource to identify key interacting functional microbial groups and to guide the design of specialized synthetic microbial communities.ResultsTo establish a system assessing comparative microbiome performance, parallel microbiomes were cultivated on sorghum (Sorghum bicolor L. Moench) from compost inocula. Biomass loss and biochemical assays indicated that these microbiomes diverged in their ability to deconstruct biomass. Network reconstructions from gene …

O6-Methylguanine-DNA-methyltransferase (MGMT) promoter methylation was shown in many studies to be an important predictive biomarker for temozolomide (TMZ) resistance and poor progression-free survival in glioblastoma multiforme (GBM) patients. However, identifying the MGMT methylation status using molecular techniques remains challenging due to technical limitations, such as the inability to obtain tumor specimens, high prices for detection, and the high complexity of intralesional heterogeneity. To overcome these difficulties, we aimed to test the feasibility of using a novel radiomics-based machine learning (ML) model to preoperatively and noninvasively predict the MGMT methylation status. In this study, radiomics features extracted from multimodal images of GBM patients with annotated MGMT methylation status were downloaded from The Cancer Imaging Archive (TCIA) public database for …

Di-(2-ethylhexyl) phthalate (DEHP) represents the most used phthalate plasticizer with an annual production above the millions of tons worldwide. Due to its inadequate disposal, outstanding chemical stability, and extremely low solubility (3 mg/L), endocrine-disrupting DEHP often accumulates in urban estuarine sediments at concentrations above the predicted no-effect concentration (20–100 mg/kg). Our previous study suggested that microbial DEHP degradation in estuarine sediments proceeds synergistically where DEHP side-chain hydrolysis to form phthalic acid represents a bottleneck. Here, we resolved this bottleneck and deconstructed the microbial synergy in O2-fluctuating estuarine sediments. Metagenomic analysis and RNA sequencing suggested that orthologous genes encoding extracellular DEHP hydrolase NCU65476 in Acidovorax sp. strain 210-6 are often flanked by the co-expressed composite …

BackgroundPredicting which pathogens might exhibit antimicrobial resistance (AMR) based on genomics data is one of the promising ways to swiftly and precisely identify AMR pathogens. Currently, the most widely used genomics approach is through identifying known AMR genes from genomic information in order to predict whether a pathogen might be resistant to certain antibiotic drugs. The list of known AMR genes, however, is still far from comprehensive and may result in inaccurate AMR pathogen predictions. We thus felt the need to expand the AMR gene set and proposed a pan-genome-based feature selection method to identify potential gene sets for AMR prediction purposes.ResultsBy building pan-genome datasets and extracting gene presence/absence patterns from four bacterial species, each with more than 2000 strains, we showed that machine learning models built from pan-genome data can be …

Background and ObjectivesShort-chain fatty acids (SCFAs) are gut microbial metabolites that promote the disease process in a rodent model of Parkinson disease (PD), but fecal levels of SCFAs in patients with PD are reduced. Simultaneous assessments of fecal and plasma SCFA levels, and their interrelationships with the PD disease process, are scarce. We aimed to compare fecal and plasma levels of different SCFA subtypes in patients with PD and healthy controls to delineate their interrelations and link to gut microbiota changes and clinical severity of PD.MethodsA cohort of 96 patients with PD and 85 controls were recruited from National Taiwan University Hospital. Fecal and plasma concentrations of SCFAs were measured using chromatography and mass spectrometry. Gut microbiota was analyzed using metagenomic shotgun sequencing. Body mass index and medical comorbidities were evaluated and …

Plant cell walls are interwoven structures recalcitrant to degradation. Both native and adapted microbiomes are particularly effective at plant cell wall deconstruction. Studying these deconstructive microbiomes provides an opportunity to assess microbiome performance and relate it to specific microbial populations and enzymes. To establish a system assessing comparative microbiome performance, parallel microbiomes were cultivated on sorghum (Sorghum bicolor L. Moench) from compost inocula. Biomass loss and biochemical assays indicated that these microbiomes diverged in their ability to deconstruct biomass. Network reconstructions from time-dependent gene expression identified key deconstructive groups within the adapted sorghum-degrading communities, including Actinotalea, Filomicrobium, and Gemmanimonadetes populations. Functional analysis of gene expression demonstrated that the microbiomes proceeded through successional stages that are linked to enzymes that deconstruct plant cell wall polymers. This combination of network and functional analysis highlighted the importance of celluloseactive Actinobacteria in differentiating the performance of these microbiomes.

Background: Emerging evidence suggests that gut dysbiosis contributes to Parkinson’s disease (PD) by signaling through microbial metabolites. Hippuric acid (HA), indole derivatives, and secondary bile acids are among the most common gut metabolites.Objective: To examine the relationship of systemic concentrations of these microbial metabolites associated with changes of gut microbiota, PD status, and severity of PD.Methods: We enrolled 56 patients with PD and 43 age-and sex-matched healthy participants. Motor and cognitive severity were assessed with Movement Disorder Society-Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) part III motor score and the Mini-Mental State Examination (MMSE), respectively. Plasma concentrations of targeted gut metabolites were measured with liquid chromatography-tandem mass spectrometry. Gut microbiota was analyzed with shotgun metagenomic …

BackgroundA well-known blood biomarker (soluble fms-like tyrosinase-1 [sFLT-1]) for preeclampsia, i.e., a pregnancy disorder, was found to predict severe COVID-19, including in males. True biomarker may be masked by more-abrupt changes related to endothelial instead of placental dysfunction. This study aimed to identify blood biomarkers that represent maternal-fetal interface tissues for predicting preeclampsia but not COVID-19 infection.MethodsThe surrogate transcriptome of tissues was determined by that in maternal blood, utilizing four datasets (n = 1354) which were collected before the COVID-19 pandemic. Applying machine learning, a preeclampsia prediction model was chosen between those using blood transcriptome (differentially expressed genes [DEGs]) and the blood-derived surrogate for tissues. We selected the best predictive model by the area under the receiver operating characteristic …

This protocol aimed to describe data transformation procedure of medical histories from electronic health records (EHRs) to historical rates by Kaplan-Meier (KM) estimation. The applicability is to extract features from real-world, time-varying data of EHRs, for developing but not limited to a machine learning prediction model. By this extraction technique, machine can learn medical history of a condition in each healthcare provider, as a differential quantity through time in term of affecting a future health state, without a need to access EHRs of other healthcare providers. However, this protocol needs a sufficient amount of longitudinal data from the most subjects in EHRs. The key stages consisted of time interval computation, historical rate derivation, and data transformation into historical rates.

We aimed to provide a framework that organizes internal properties of a convolutional neural network (CNN) model using non-image data to be interpretable by human. The interface was represented as ontology map and network respectively by dimensional reduction and hierarchical clustering techniques. The applicability is to implement a prediction model either to classify categorical or to estimate numerical outcome, including but not limited to that using data from electronic health records. This pipeline harnesses invention of CNN algorithms for non-image data while improving the depth of interpretability by data-driven ontology. However, the DI-VNN is only for exploration beyond its predictive ability, which requires further explanatory studies, and needs a human user with specific competences in medicine, statistics, and machine learning to explore the DI-VNN with high confidence. The key stages consisted of data preprocessing, differential analysis, feature mapping, network architecture construction, model training and validation, and exploratory analysis.

Celastrol is a quinone-methide triterpenoid isolated from the root extracts of Tripterygium wilfordii (Thunder god vine). Although celastrol possesses multiple bioactivities, the potent toxicity and rare solubility in water hinder its clinical application. Biotransformation of celastrol using either whole cells or purified enzymes to form less toxic and more soluble derivatives has been proven difficult due to its potent antibiotic and enzyme-conjugation property. The present study evaluated biotransformation of celastrol by four glycosyltransferases from Bacillus species and found one glycosyltransferase (BsGT110) from Bacillus subtilis with significant activity toward celastrol. The biotransformation metabolite was purified and identified as celastrol-29-O-β-glucoside by mass and nuclear magnetic resonance spectroscopy. Celastrol-29-O-β-glucoside showed over 53-fold higher water solubility than celastrol, while maintained 50 …

We proposed a learning algorithm for human to conduct literature and data mining for causal factor discovery. The applicability is to select features for a machine learning prediction model, including but not limited to that using real-world, time-varying data from electronic health records. This protocol is relatively quick to find potentially actionable predictors for a clinical prediction while dealing with high dimensionality in big data. However, this protocol might not find a potentially novel cause, since this only exhaustively examines the existing evidences in a single study. The key stages consisted of systematic human learning, causal diagram construction, data preprocessing, causal inference modeling, and development and validation of a prediction model to describe the explainability.

The soil bacterium Psychrobacillus sp. strain AK 1817 was isolated from a tropical soil sample collected in Taiwan. Strain AK 1817 biotransforms the ergostane triterpenoid antcin K from the fungus Antrodia cinnamomea. The genome was sequenced using the PacBio RS II platform and consists of one chromosome of 4,096,020 bp, comprising 3,907 protein-coding genes, 75 tRNAs, 30 rRNAs, 5 noncoding RNAs (ncRNAs), and 100 pseudogenes.

Background/purposeMetabolites in blood have been found associated with the occurrence of vascular diseases, but its role in the functional recovery of stroke is unclear. The aim of this study is to investigate whether the untargeted metabolomics at the acute stage of ischemic stroke is able to predict functional recovery.MethodsOne hundred and fifty patients with acute ischemic stroke were recruited and followed up for 3 months. Fasting blood samples within 7 days of stroke were obtained, liquid chromatography and mass spectrometry were applied to identify outcome-associated metabolites. The patients’ clinical characteristics and identified metabolites were included for constructing the outcome prediction model using machine learning approaches.ResultsBy using multivariate analysis, 220 differentially expressed metabolites (DEMs) were discovered between patients with favorable outcomes (modified Rankin …

The first mitochondrial genome of Ophiocara porocephala was determined by the combination of next-generation sequencing (NGS) and Sanger sequencing methods. A complete circular mitogenome of O. porocephala (16,529 bp) consisted of 13 protein-coding genes, 22 transfer RNAs, two ribosomal RNAs, and two non-coding regions, including a control region (D-loop) and a light strand origin of replication (OL). Two start codons (ATG and GTG) and four stop codons (TAG, TAA, TA–, and T–) were used in all the PCGs. Except for ND6 and eight transfer RNAs (tRNAs), all the other genes were encoded in the heavy strand. Based on phylogenetic analysis, O. porocephala formed a clade with three other species in the subfamily Butinae, while the other 10 made a subfamily Eleotrinae clade.

ImportancePrognostic predictions of prelabor rupture of membranes lack proper sample sizes and external validation.ObjectiveTo develop, validate, and deploy statistical and/or machine learning prediction models using medical histories for prelabor rupture of membranes and the time of delivery.DesignA retrospective cohort design within 2-year period (2015 to 2016) of a single-payer, government-owned health insurance database covering 75.8% individuals in a countrySettingNationwide healthcare providers (n=22,024) at primary, secondary, and tertiary levelsParticipants12-to-55-year-old women that visit healthcare providers using the insurance from ∼1% random sample of insurance holders stratified by healthcare provider and category of family: (1) never visit; (2) visit only primary care; and (3) visit all levels of carePredictorsMedical histories of diagnosis and procedure (International Classification of Disease version 10) before the latest visit of outcome within the database periodMain Outcomes and MeasuresPrelabor rupture of membranes prognostication (area under curve, with sensitivity, specificity, and likelihood ratio), the time of delivery estimation (root mean square error), and inference time (minutes), with 95% confidence intervalResultsWe selected 219,272 women aged 33 ± 12 years. The best prognostication achieved area under curve 0.73 (0.72 to 0.75), sensitivity 0.494 (0.489 to 0.500), specificity 0.816 (0.814 to 0.818), and likelihood ratio being positive 2.68 (2.63 to 2.75) and negative 0.62 (0.61 to 0.63). This outperformed models from previous studies according to area under curve of an external validation set, including one …

It is critical to identify individual genomes from microbiomic samples in order to carry out analysis of the microbes. Methods based on existing databases, however, may have limited capabilities in elucidating and quantifying the microbes due to the largely unidentified microbial species in natural or human‐associated environments. We thus developed a database‐free method, MaxBin 2.0, to aid in the process of recovering microbial genomes from metagenomes in a de novo manner. The recovery of individual genomes allows analysis of the microbiome in terms of a collection of microbial genomes so that one can understand the functional roles of each species. The data of individual microbes may then be analyzed collectively to untangle the interactions between different microbial organisms. By reporting the genome abundance information for co‐assembled metagenomes, one may also identify which …

Background Discerning genes crucial to antimicrobial resistance (AMR) mechanisms is becoming more and more important to accurately and swiftly identify AMR pathogenic strains. Pangenome-wide association studies (e.g. Scoary) identified numerous putative AMR genes. However, only a tiny proportion of the putative resistance genes are annotated by AMR databases or Gene Ontology. In addition, many putative resistance genes are of unknown function (termed hypothetical proteins). An annotation tool is crucially needed in order to reveal the functional organization of the resistome and expand our knowledge of the AMR gene repertoire. Results We developed an approach (PangenomeNet) for building co-functional networks from pan-genomes to infer functions for hypothetical genes. Using Escherichia coli as an example, we demonstrated that it …

Ganoderma lucidum is a medicinal fungus whose numerous triterpenoids are its main bioactive constituents. Although hundreds of Ganoderma triterpenoids have been identified, Ganoderma triterpenoid glycosides, also named triterpenoid saponins, have been rarely found. Ganoderic acid A (GAA), a major Ganoderma triterpenoid, was synthetically cascaded to form GAA-15-O-β-glucopyranoside (GAA-15-G) by glycosyltransferase (BtGT_16345) from Bacillus thuringiensis GA A07 and subsequently biotransformed into a series of GAA glucosides by cyclodextrin glucanotransferase (Toruzyme® 3.0 L) from Thermoanaerobacter sp. The optimal reaction conditions for the second-step biotransformation of GAA-15-G were found to be 20% of maltose; pH 5; 60 °C. A series of GAA glucosides (GAA-G2, GAA-G3, and GAA-G4) could be purified with preparative high-performance liquid chromatography (HPLC) and identified by mass and nucleic magnetic resonance (NMR) spectral analysis. The major product, GAA-15-O-[α-glucopyranosyl-(1→4)-β-glucopyranoside] (GAA-G2), showed over 4554-fold higher aqueous solubility than GAA. The present study demonstrated that multiple Ganoderma triterpenoid saponins could be produced by sequential actions of BtGT_16345 and Toruzyme®, and the synthetic strategy that we proposed might be applied to many other Ganoderma triterpenoids to produce numerous novel Ganoderma triterpenoid saponins in the future.

This protocol aims to develop, validate, and deploy a prediction model using high dimensional data by both human and machine learning. The applicability is intended for clinical prediction in healthcare providers, including but not limited to those using medical histories from electronic health records. This protocol applies diverse approaches to improve both predictive performance and interpretability while maintaining the generalizability of model evaluation. However, some steps require expensive computational capacity; otherwise, these will take longer time. The key stages consist of designs of data collection and analysis, feature discovery and quality control, and model development, validation, and deployment.

Backgrounds: Influenza vaccination could decrease the risk of major cardiac events in patients with chronic obstructive pulmonary disease (COPD). However, the vaccine’s effects on decreasing the risk of ventricular arrhythmia (VA) development in such patients remain unclear. Methods: We retrospectively analysed the data of 18,658 patients with COPD (≥55 years old) from the National Health Insurance Research Database during January 1, 2001 to December 31, 2012. After a 1:1 propensity score matching by the year of diagnosis, we divided the patients into vaccinated and unvaccinated groups. Time-varying Cox proportional hazards regression was applied to assess the time to event hazards of influenza vaccination exposure Results: The risk of VA occurrence was significantly lower in the vaccinated group during influenza season and all seasons (adjusted hazard ratio [aHR]: 0.62, 95% confidence interval [CI]: 0.41–0.95; aHR: 0.69, 95% CI: 0.44–1.08; and aHR: 0.65, 95% CI: 0.48–0.89, in influenza season, non-influenza season and all seasons respectively). Among patients with CHA2DS2-VASc scores of 2-3, receiving one time and two to three times of influenza vaccination were associated with lower risk of VA occurrence in all seasons (aHR: 0.28, 95% CI: 0.10–0.80; aHR: 0.27, 95% CI: 0.10–0.68, respectively). Among patients without stroke, peripheral vascular disease and diabetes, lower risk of VA occurrence after receiving one and two to three times of vaccination was observed in all seasons. Among patients with history of asthma and patients without history of heart failure, ischemic heart disease, angina hypertension or renal …

Di-(2-ethylhexyl) phthalate (DEHP) is the most widely used plasticizer worldwide, with an annual global production of more than 8 million tons. Because of its improper disposal, endocrine-disrupting DEHP often accumulates in estuarine sediments in industrialized countries at submillimolar levels, resulting in adverse effects on both ecosystems and human beings. The microbial degraders and biodegradation pathways of DEHP in O2-limited estuarine sediments remain elusive. Here, we employed an integrated meta-omics approach to identify the DEHP degradation pathway and major degraders in this ecosystem. Estuarine sediments were treated with DEHP or its derived metabolites, o-phthalic acid and benzoic acid. The rate of DEHP degradation in denitrifying mesocosms was two times slower than that of o-phthalic acid, suggesting that side chain hydrolysis of DEHP is the rate-limiting step of anaerobic DEHP …

Background: Preeclampsia and intrauterine growth restriction are placental dysfunction–related disorders (PDDs) that require a referral decision be made within a certain time period. An appropriate prediction model should be developed for these diseases. However, previous models did not demonstrate robust performances and/or they were developed from datasets with highly imbalanced classes.Objective: In this study, we developed a predictive model of PDDs by machine learning that uses features at 24-37 weeks’ gestation, including maternal characteristics, uterine artery (UtA) Doppler measures, soluble fms-like tyrosine kinase receptor-1 (sFlt-1), and placental growth factor (PlGF).Methods: A public dataset was taken from a prospective cohort study that included pregnant women with PDDs (66/95, 69%) and a control group (29/95, 31%). Preliminary selection of features was based on a statistical analysis using SAS 9.4 (SAS Institute). We used Weka (Waikato Environment for Knowledge Analysis) 3.8. 3 (The University of Waikato, Hamilton, NZ) to automatically select the best model using its optimization algorithm. We also manually selected the best of 23 white-box models. Models, including those from recent studies, were also compared by interval estimation of evaluation metrics. We used the Matthew correlation coefficient (MCC) as the main metric. It is not overoptimistic to evaluate the performance of a prediction model developed from a dataset with a class imbalance. Repeated 10-fold cross-validation was applied.Results: The classification via regression model was chosen as the best model. Our model had a robust MCC (. 93, 95 …

The capability of gut microbiota in degrading foods and drugs administered orally can result in diversified efficacies and toxicity interpersonally and cause significant impact on human health. Production of atherogenic trimethylamine N-oxide (TMAO) from carnitine is a gut microbiota-directed pathway and varies widely among individuals. Here, we demonstrated a personalized TMAO formation and carnitine bioavailability from carnitine supplements by differentiating individual TMAO productivities with a recently developed oral carnitine challenge test (OCCT). By exploring gut microbiome in subjects characterized by TMAO producer phenotypes, we identified 39 operational taxonomy units that were highly correlated to TMAO productivity, including Emergencia timonensis, which has been recently discovered to convert γ-butyrobetaine to TMA in vitro. A microbiome-based random forest classifier was …

Pattern recognition has been widely used in various applications of image processing. It is used to extract meaningful image features from the given image samples and to build classification systems with the intelligence of human recognition. Convolutional Neural Network (CNN) [1] has been one of the most popular and widely used methods for image pattern recognition applications. However, CNN was known not to be rotation-invariant to image patterns. It usually required a larger amount of training image dataset with greater variations in positions and orientations, or additional numerical treatments of spatial transformations [2]. On the other hand, K-mer-based Pattern Recognition (KPR) [3] has been developed to apply an unique way of rotation-invariant sampling to the inspected image pattern and analyze the frequency of the captured pattern features. A classification system was then built based on the K …

Morphine is a strong painkiller acting through mu-opioid receptor (MOR). Full-length 7-transmembrane (TM) variants of MOR share similar amino acid sequences of TM domains in rodents and humans; however, interspecies differences in N-and C-terminal amino acid sequences of MOR splice variants dramatically affect the downstream signaling. Thus, it is essential to develop a mouse model that expresses human MOR splice variants for opioid pharmacological studies. We generated 2 lines of fully humanized MOR mice (hMOR 1; mMOR 2/2 mice), line# 1 and# 2. The novel murine model having human OPRM1 genes and human-specific variants was examined by reverse-transcription polymerase chain reaction and the MinION nanopore sequencing. The differences in the regional distribution of MOR between wild-type and humanized MOR mice brains were detected by RNAscope and radioligand binding …

Background: Predictions in pregnancy care are complex because of interactions among multiple factors. Hence, pregnancy outcomes are not easily predicted by a single predictor using only one algorithm or modeling method.Objective: This study aims to review and compare the predictive performances between logistic regression (LR) and other machine learning algorithms for developing or validating a multivariable prognostic prediction model for pregnancy care to inform clinicians’ decision making.Methods: Research articles from MEDLINE, Scopus, Web of Science, and Google Scholar were reviewed following several guidelines for a prognostic prediction study, including a risk of bias (ROB) assessment. We report the results based on the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. Studies were primarily framed as PICOTS (population, index, comparator, outcomes, timing, and setting): Population: men or women in procreative management, pregnant women, and fetuses or newborns; Index: multivariable prognostic prediction models using non-LR algorithms for risk classification to inform clinicians’ decision making; Comparator: the models applying an LR; Outcomes: pregnancy-related outcomes of procreation or pregnancy outcomes for pregnant women and fetuses or newborns; Timing: pre-, inter-, and peripregnancy periods (predictors), at the pregnancy, delivery, and either puerperal or neonatal period (outcome), and either short-or long-term prognoses (time interval); and Setting: primary care or hospital. The results were synthesized by reporting study characteristics and ROBs and by …