PubMed Health⌕ Search

Biomedical subjects

Yifan Lin

Publications and source records attributed to Yifan Lin.

2 recordsLinked to original sources
PubMed2026

NS2A V89F mutation in a DENV1 clinical isolate enhances neurotropism and neuroinvasion.

INTRODUCTION: Dengue virus (DENV) neurological complications are increasingly reported, yet the viral genetic determinants of neurotropism remain poorly characterized. METHODS: We screened 25 DENV1 clinical isolates from the 2014 outbreak in Guangdong, China, for neurotropism in suckling mice, and integrated comparative genomics, pre-expression functional assays, population-scale sequence analysis, and OpenFold3 structural modeling to identify mutations associated with enhanced neuroinvasion. RESULTS: We found that only strain P1253 induced neurological symptoms and mortality via subcutaneous inoculation, producing cortical-selective lesions distinct from the diffuse encephalitic damage observed after intracranial inoculation, and P1253 replicated preferentially in human brain microvascular endothelial cells (HBMEC) compared to contemporaneous strains. Comparative genomics identified three unique mutations in P1253 (NS1 175Y→H, NS2A 89V→F, NS4A 2V→I), and pre-expression assays demonstrated that only NS2A 89V→F significantly enhanced viral replication and cytopathic effect in HBMEC. Analysis of 1,990 complete DENV1 genomes revealed five natural mutant types in the NS2A 89 -96 residue region, with P1253 representing the FIPI quadruple-mutant type, and OpenFold3 structural prediction showed that 89V→F introduced on the VIPI background induced the most significant distal domain reorientation (RMSD 1.605 Å), increasing the centroid-to-centroid distance between residues 89 -96 and 185 -218 from 18.221 Å to 27.462 Å. DISCUSSION: These findings identify NS2A 89V→F as a candidate adaptive mutation associated with enhanced neurotropism in DENV1 and provide a framework for monitoring neurovirulent variants.

Dengue Virus↗
PubMed2026

High-Dimensional Sensitivity Analysis for Genomic Studies: An Adversarial Framework for Learning Worst-Case Latent Confounders.

High-dimensional genomics studies are frequently confounded by unmeasured biological processes that obscure disease-specific signals. While existing workflows can estimate these latent confounders, they fail to quantify how robust a discovery is to varying levels of hypothetical confounding. We introduce sensGAN, a deep-learning adversarial framework that systematically explores the confounding spectrum by learning "worst-case" latent variables that nullify the most gene associations under novel predictive-gain constraints. By identifying the minimum confounding strength required to explain away an observed effect, our method shifts the paradigm toward a formal, quantitative sensitivity analysis. In diverse simulations, sensGAN accurately recovers latent structures and outperforms existing methods in identifying confounder-sensitive genes. Applied to human Alzheimer's disease microglia, our framework prioritizes robust disease pathways while successfully isolating signals driven by unmeasured co-occurring neurodegenerative pathologies. Our method is publicly available, deposited at the GitHub repository yifanlinz/ADsensitivityICML.

Journal Article↗