Abstract Manual curation of gene–disease–phenotype relationships from the human genetics literature is a persistent bottleneck for maintaining its bioinformatics databases. Whereas large language models (LLMs) offer a promising alternative, there is currently no systematic benchmark that evaluates whether state‐of‐the‐art commercial LLMs can perform this task reliably on the full‐text articles. To address this gap, we introduce a standardized benchmark comprising 406 full‐text articles covering 180 congenital heart disease‐associated genes, and a multi‐dimensional evaluation framework that incorporates fuzzy matching to account for synonyms and partial matches. We benchmarked seven state‐of‐the‐art LLMs, GPT‐4o, Claude‐Opus‐4, DeepSeek‐R1, Grok‐4, Qwen‐3.5, Gemini‐2.5 (Pro), and GPT‐5 on the extraction of structured gene, disease, and phenotype fields. The top‐performing model, Grok‐4, achieved 97.6% overall accuracy, whereas the lowest‐performing model reached approximately 88%, still surpassing many prior benchmarks employing zero‐shot or n‐shot prompting in biomedical relation extraction (RE) tasks. Our results provide a rigorous characterization of current LLMs capabilities and limitations. This paper contains two components. First, we conducted a human genetics field benchmark study on LLMs against a curated database. Second we developed the evaluation framework for this task. The benchmark dataset, evaluation framework, and model benchmarking outputs are made available online to support future studies in a reproducible manner.
Abstract The evolution of severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) during the coronavirus disease 2019 (COVID‐19) pandemic highlights the critical need for predicting virus mutations in order to stay ahead of infectious diseases. Here, we present the time‐enhanced attention‐based model (TEAM), which combines phylogenetic sampling with deep learning to improve virus mutation prediction accuracy. TEAM introduces a novel time‐enhanced phylogenetic sampling strategy that preserves both evolutionary and temporal sequence relationships, enhancing its ability to predict site‐specific mutations as a multi‐class classification task. The framework leverages evolutionary scale modeling embeddings and a dual‐attention mechanism to capture sequence‐level patterns and temporal dynamics. Experiments on the SARS‐CoV‐2 spike protein dataset show that TEAM substantially outperforms existing methods. Further evaluations on membrane protein and H1N1 datasets confirm its robustness and generalizability. In practice, TEAM provides a scalable and interpretable solution for mutation prediction, offering valuable insights for evolutionary research and public health planning.
Abstract In this study, a fractional‐order SEIuITR model was proposed to examine the transmission of tuberculosis (TB) in the community. The proposed model was rigorously examined for well‐posedness. The basic reproduction number was also derived, and the disease‐free equilibrium was obtained. The local and global stabilities of the proposed model were also examined. The proposed model was fitted with the actual TB incidence in Ethiopia, while numerical simulations were carried out in MATLAB to validate the results obtained in this study. Local and global sensitivity analyses were carried out to determine the key parameters that influence the transmission of TB in the community. Phase‐plane portraits and phase‐plane diagrams demonstrate the dynamic process of the state variables. The results indicate that as the fractional‐order increases, the number of infections decreases due to the memory effect of early detection and treatment, while as the order decreases, the number of infections increases. Early detection and treatment have a profound effect on recovery and prevalence.
Abstract Single‐cell RNA sequencing (scRNA‐seq) has transformed the study of cellular heterogeneity by making it possible to classify individual cells and their functional states. However, the analysis remains difficult because high dropout rates lead to sparse and noisy expression data. Existing marker gene selection methods are often fragmented, typically relying on a single strategy that overlooks complementary biological signals. Without a unifying framework for denoising and marker identification, clustering can suffer in accuracy, stability, and interpretability. This makes it more difficult to define cell subpopulations and extract meaningful biological insights clearly. We introduce single cell marker gene clustering (scMGC), a unified framework that integrates a denoising autoencoder with multi‐method marker selection for scRNA‐seq clustering. The autoencoder reduces dropout noise, whereas a unified marker gene scoring system balances the contributions of multiple methods to select reliable markers. These markers are then applied in graph‐based clustering to uncover accurate and interpretable cell subpopulations. scMGC was benchmarked against seven state‐of‐the‐art methods across seven scRNA‐seq datasets. On average, it outperformed competing approaches by 31.3% in adjusted rand index and 28.2% in normalized mutual information, showing consistent improvements in clustering accuracy. Enrichment and disease association analyses further validated that the discovered clusters are both biologically meaningful and clinically relevant. The codes and datasets used are available on the GitHub website (srniloy/scMGC).
