Jianchao Ying

Efflux proteins (EPs) are key transporters responsible for expelling antibiotics and other harmful substances from bacterial cells, thus contributing significantly to antimicrobial resistance. As such, the development of a reliable and efficient computational method for identifying EPs in bacteria and archaea is of utmost importance. However, current computational methods for EP identification are limited and suffer from various performance and utility constraints. This study presents EPDiscovery, a stacking-based ensemble classifier designed to predict prokaryotic EPs directly from their amino acid sequence. EPDiscovery employs the MLP algorithm to integrate the predictions of fifteen baseline classifiers. The test results demonstrated that EPDiscovery accurately identified EPs, achieving AUROC, accuracy, F-value, and Matthews correlation coefficient scores of 0.970, 0.926, 0.927, and 0.852, respectively. Additionally, EPDiscovery outperformed the existing representative method and demonstrated consistent and satisfactory performance, with AUROC scores exceeding 0.9 across three previously proposed datasets. Feature importance analysis using local interpretable model-agnostic explanations further confirmed EPDiscovery’s ability to discern the characteristics of EPs from protein sequence information. EPDiscovery offers a reliable alternative for annotating efflux proteins in large-scale bacterial genome studies, thus accelerating the discovery of potential EPs and contributing to the advancement of antimicrobial resistance research.

Background: Thalassemia is a common hereditary anemia that poses diagnostic challenges during pregnancy due to physiological changes in hematological parameters. Misdiagnosis or delayed diagnosis can result in serious maternal and fetal complications. Although genetic testing is the gold standard for thalassemia diagnosis, its high cost and limited availability restrict its use in primary care settings. This study aimed to develop and validate a clinically interpretable machine learning model to identify thalassemia in pregnant women using routine complete blood count indicators. Methods: A total of 523 pregnant patients were retrospectively enrolled in this study. Eight predictive features were selected using the Boruta algorithm. Multiple machine learning algorithms were trained and evaluated using ten-fold cross-validation. An ensemble model was constructed through bagging and weighted integration of top-performing models. Model performance was assessed using a series of metrics such as AUROC, AUPRC, accuracy, F1-score, and MCC. Comparative evaluation was conducted against traditional diagnostic indices and a recently proposed model. Model interpretability was analyzed using permutation importance and SHAP methods. Results: The Thal-classifier, constructed by integrating four bagged machine learning models, was developed for accurate identification of thalassemia. It exhibited excellent diagnostic performance, with an AUROC of 0.945 on the cross-validation and 0.899 on the independent testing dataset. Compared with the existing indices and models, it achieved superior performance across multiple evaluation metrics. Feature importance analyses consistently identified RDW-SD and MCHC as the primary predictive features. Conclusion: Thal-classifier is a practical and reliable tool for early detection of thalassemia in pregnant populations, offering valuable support for prenatal screening and individualized clinical decision-making.

Background: Circulating microRNAs (miRNAs) are recently a rapidly increasing of interest as non-invasive biomarkers for diagnosis and prognosis of acute myocardial infarction (AMI). Previous studies revealed that several miRNAs exhibited the capacity for diagnosis and prognosis of AMI, the reasons why these circulating miRNAs are concerned as targets for investigation are quite cryptogenic, presumably due to the lack of clues provided by global surveillance at the transcriptome level, and the current data for some miRNAs are controversial and inconsistent among independent studies. Methods: To comprehensively profiling the potential miRNAs for diagnosis and prognosis of AMI, we reported transcriptomes of circulating miRNAs in the plasma of 27 healthy controls, 64 AMI patients (37 STEMI and 27 NSTEMI) and 20 AMI patients who were subjected to reperfusion therapy. Meanwhile, the cTnI of AMI patients was parallel determined. Differentially-circulated miRNAs were analyzed between each group. All detected circulating miRNAs were examined by ROC analysis and then LASSO dimension reduction to obtain an optimal panel for diagnosis of AMI. A five-year period follow-up towards the AMI and reperfusion patients was performed, and the prognostic value of circulating miRNAs in these patients was estimated by using the Cox regression model, ROC and Kaplan-Meier curves. Results: Comprehensive global differences of miRNAs transcriptome among AMI, reperfusion patients and healthy controls were identified. A total of 40 miRNAs, called high diagnostic performance miRNAs, including several previous well-studied miRNAs with AUC greater than 0.85 were shown to discriminate AMI with healthy controls. In addition, 29 miRNAs were analyzed to be strongly correlated with the plasma cTnI level, of which 20 overlapped with high diagnostic performance miRNAs. These overlapped miRNAs are over-represented in the pathways which actually reflect the pathological cause of myocardial infarction, as well as the regulation of gene expression and energetic pathway of cellular response to hypoxia. Finally, two miRNAs were analyzed to be significantly correlated to all-cause mortality. Conclusion: This is the first time to survey plasma miRNAs for the development of AMI diagnostic and prognostic biomarkers at the transcriptome level. A subset of miRNAs exhibited potential diagnostic and prognostic merits for AMI.

Identification of prokaryotic transposases (Tnps) not only gives insight into the spread of antibiotic resistance and virulence but the process of DNA movement. This study aimed to develop a classifier for predicting Tnps in bacteria and archaea using machine learning (ML) approaches. We extracted a total of 2751 protein features from the training dataset including 14852 Tnps and 14852 controls, and selected 75 features as predictive signatures using the combined mutual information and least absolute shrinkage and selection operator algorithms. By aggregating these signatures, an ensemble classifier that integrated a collection of individual ML-based classifiers, was developed to identify Tnps. Further validation revealed that this classifier achieved good performance with an average AUC of 0.955, and met or exceeded other common methods. Based on this ensemble classifier, a stand-alone command-line tool designated TnpDiscovery was established to maximize the convenience for bioinformaticians and experimental researchers toward Tnp prediction. This study demonstrates the effectiveness of ML approaches in identifying Tnps, facilitating the discovery of novel Tnps in the future.

As endometrial cancer (EC) is a major threat to female health worldwide, the ability to provide an accurate diagnosis and prognosis of EC is promising to improve its treatment guidance. Since the discovery of miRNAs, it has been realized that miRNAs are associated with every cell function, including malignant transformation and metastasis. This study aimed to explore diagnostic and prognostic miRNA markers of EC. In this study, differential analysis and machine learning were performed, followed by correlation analysis of miRNA‐mRNA based on the miRNA and mRNA expression data. Nine miRNAs were identified as diagnostic markers, and a diagnostic classifier was established to distinguish between EC and normal endometrium tissue with overall correct rates >95%. Five specific prognostic miRNA markers were selected to construct a prognostic model, which was confirmed more effective in identifying EC patients at high risk of mortality compared with the FIGO staging system. This study demonstrates that the expression patterns of miRNAs may hold promise for becoming diagnostic and prognostic biomarkers and novel therapeutic targets for EC.

Rhodococcus equi, a member of the Rhodococcus genus, is a gram-positive pathogenic bacterium. Rhodococcus possesses an open pan-genome that constitutes the basis of its high genomic diversity and allows for adaptation to specific niche conditions and the changing host environments. Our analysis further showed that the core genome of R. equi contributes to the pathogenicity and niche adaptation of R. equi. Comparative genomic analysis revealed that the genomes of R. equi shared identical collinearity relationship, and heterogeneity was mainly acquired by means of genomic islands and prophages. Moreover, genomic islands in R. equi were always involved in virulence, resistance, or niche adaptation and possibly working with prophages to cause the majority of genome expansion. These findings provide an insight into the genomic diversity, evolution, and structural variation of R. equi and a valuable resource for functional genomic studies.

Wilm's tumour‐1 (WT1) is overexpressed in pancreatic ductal adenocarcinoma (PDAC) and enhances metastasis. Deubiquitination stabilizes target proteins, and inhibiting deubiquitination facilitates the degradation of target proteins. However, whether inhibiting deubiquitination of WT1 facilitates its degradation and presents anti‐cancer ability in PDAC is unknown. Here, we found that deubiquitinase inhibitor degrasyn rapidly induced the degradation of endogenous and exogenous WT1 through enhancing ubiquitination of WT1 followed by the up‐regulation of E‐cadherin. Knockdown of WT1 by short hairpin RNAs (shRNAs) inhibited metastasis and overexpression of WT1 partially prevented degrasyn‐induced anti‐metastasis activity, suggesting that degrasyn presents anti‐metastasis activity partially through degrading WT1 protein. We further identified that USP5 deubiquitinated WT1 and stabilized its expression. The higher expressions of USP5 and WT1 are associated with tumour metastasis. More importantly, degrasyn inhibited the activity of USP5 and overexpression of USP5 partially prevented degrasyn‐induced degradation of WT1 protein, suggesting that degrasyn degraded WT1 protein through inhibiting the activity of USP5. Finally, degrasyn reduced the tumorigenicity in a xenograft mouse model and reduced the metastasis in vivo. Our results indicate that degrasyn presents strong anti‐cancer activity through USP5‐WT1‐E‐cadherin signalling in PDAC. Therefore, degrasyn holds promise as cancer therapeutic agent in PDAC with high expressions of USP5 and WT1.