Biomedical subjects

Guo Wei

Publications and source records attributed to Guo Wei.

3 recordsLinked to original sources

PubMed2026

Metagenomic next-generation sequencing for tuberculosis diagnosis: enhanced performance and cost-effectiveness.

UNLABELLED: Metagenomic next-generation sequencing (mNGS) is a promising tool for diagnosing challenging infections like tuberculosis (TB). However, previous studies largely focused on case-specific application of mNGS in TB diagnosis. Thus, we conducted a retrospective observational study to first systematically evaluate the diagnostic performance and cost-effectiveness of mNGS for TB diagnosis. We retrieved a total of 16,776 results of the seven TB diagnostic assays, including mNGS, tuberculosis IgG antibody, TB interferon-γ release assay (TB-IGRA), TB-DNA, Xpert MTB/RIF (Xpert), culture, and acid-fast bacilli staining (AFS) from 3,757 participants with suspected TB infection at Sichuan Provincial People's Hospital from September 2021 to July 2024. Diagnostic metrics were compared against a composite reference standard. Microbial composition and a cost-utility analysis were performed. Among seven TB assays studied, the World Health Organization (WHO)-recommended assays AFS, culture, and Xpert, as well as TB-IGRA, were requested most frequently for TB diagnosis, whereas mNGS ranked last. mNGS demonstrated the highest specificity (100%), accuracy (72.3%), and area under the curve (AUC) (0.795). Its sensitivity in bronchoalveolar lavage fluid and tissue was 71.0% and 72.7%, respectively. Sequential use of mNGS after initial WHO-recommended tests (Xpert/Culture/AFS) significantly improved diagnostic performance (sensitivity, 70.4%; AUC, 0.823). Microbial analysis associated Candida albicans with TB. Cost-utility analysis showed sequential mNGS became cost-effective at higher willingness-to-pay thresholds (>200,000 RMB per correct diagnosis). mNGS offers superior specificity for TB diagnosis. A sequential strategy applying mNGS to conventional-test-negative cases provides enhanced diagnostic performance and is cost-effective at higher healthcare investment values, supporting its utility for diagnostically challenging TB. IMPORTANCE: This study systematically assesses the diagnostic performance and cost utility of metagenomic next-generation sequencing (mNGS) for tuberculosis (TB) in a large real-world cohort of 3,757 suspected patients, comparing it against six conventional assays (tuberculosis IgG antibody, TB interferon-γ release assay, TB-DNA, Xpert, culture, and acid-fast bacilli staining). mNGS demonstrated the highest specificity (100%), accuracy (72.3%), and area under the curve (AUC) (0.795), with sensitivities of 71.0% in bronchoalveolar lavage fluid and 72.7% in tissue. Notably, sequential use of mNGS after the World Health Organization-recommended tests significantly improved sensitivity to 70.4% and AUC to 0.823. Candida albicans showed significant differences among the three groups. The sequential mNGS strategy was cost-effective compared with no mNGS, and its cost-effectiveness increased with a rising willingness-to-pay threshold. Overall, these results highlight mNGS as a valuable supplementary tool for challenging TB cases, especially when conventional tests are inconclusive, and provide strong evidence for integrating it into diagnostic algorithms to optimize clinical decision-making and resource allocation.

Adult↗

PubMed2026

Genome- and peak-informed two-stage framework for scATAC-seq cell type identification.

MOTIVATION: Accurate cell type annotation is essential in scATAC-seq analysis, as it underpins the characterization of cellular heterogeneity, the identification of regulatory elements, and downstream biological discovery. However, current annotation methods still face major challenges. First, although some approaches attempt to integrate genomic sequence information, they typically rely on shallow sequence representations and thus fail to capture the long-range dependencies and regulatory signals encoded in DNA. Second, substantial batch effects introduced by different platforms, sequencing batches, or tissue sources remain insufficiently addressed. Existing models often lack robust distribution alignment and domain generalization capabilities, leading to confounding non-biological variation and reduced annotation accuracy across datasets. RESULTS: To overcome these limitations, we propose seqAlignATAC, a two-stage intra-modality annotation framework that integrates sequence-derived embeddings with domain adaptation. In the first stage, we employ a large-scale pretrained nucleotide language model to extract low-dimensional, biologically informative representations from the genomic sequences of chromatin-accessible peaks. In the second stage, these embeddings are fed into a supervised neural network equipped with an adaptive alignment module to mitigate batch effects and harmonize feature distributions between labeled reference and unlabeled target datasets. Extensive experiments across multiple settings demonstrate that seqAlignATAC achieves competitive accuracy and robustness, effectively leveraging genome-level information while alleviating batch-induced distributional discrepancies. AVAILABILITY AND IMPLEMENTATION: The source code of seqAlignATAC is available at: https://github.com/BioCS-Lab/seqAlignATAC.

Humans↗

PubMed2025

A Graph Contrastive Learning Method for Enhancing Genome Recovery in Complex Microbial Communities.

Accurate genome binning is essential for resolving microbial community structure and functional potential from metagenomic data. However, existing approaches-primarily reliant on tetranucleotide frequency (TNF) and abundance profiles-often perform sub-optimally in the face of complex community compositions, low-abundance taxa, and long-read sequencing datasets. To address these limitations, we present MBGCCA, a novel metagenomic binning framework that synergistically integrates graph neural networks (GNNs), contrastive learning, and information-theoretic regularization to enhance binning accuracy, robustness, and biological coherence. MBGCCA operates in two stages: (1) multimodal information integration, where TNF and abundance profiles are fused via a deep neural network trained using a multi-view contrastive loss, and (2) self-supervised graph representation learning, which leverages assembly graph topology to refine contig embeddings. The contrastive learning objective follows the InfoMax principle by maximizing mutual information across augmented views and modalities, encouraging the model to extract globally consistent and high-information representations. By aligning perturbed graph views while preserving topological structure, MBGCCA effectively captures both global genomic characteristics and local contig relationships. Comprehensive evaluations using both synthetic and real-world datasets-including wastewater and soil microbiomes-demonstrate that MBGCCA consistently outperforms state-of-the-art binning methods, particularly in challenging scenarios marked by sparse data and high community complexity. These results highlight the value of entropy-aware, topology-preserving learning for advancing metagenomic genome reconstruction.

canonical correlation analysis↗