检索结果-内蒙古大学图书馆

Analyzing COVID-19 progression with Markov multistage models: insights from a Korean cohort

Genomics and Informatics 2025年第1期23卷 1-11页

作者： Ndagijimana, Frank Aimee Rodrigue Park, Taesung Interdisciplinary Program in Bioinformatics Seoul National University Seoul South Korea Department of Statistics Seoul National University Seoul South Korea

Background: Understanding the progression and recovery process of COVID-19 is crucial for guiding public health strategies and developing targeted interventions. This longitudinal cohort study aims to elucidate the dynamics of COVID-19 severity progression and evaluate the impact of underlying health conditions on these transitions, providing critical insights for more effective disease management. Methods: Data from 4549 COVID-19 patients admitted to Seoul National University Boramae Medical Center between February 5th, 2020, and October 30th, 2021, were analyzed using a 5-state continuous-time Markov multistate model. The model estimated instantaneous transition rates between different levels of COVID-19 severity, predicted probabilities of state transitions, and determined hazard ratios associated with underlying comorbidities. Results: The analysis revealed that most patients stabilized in their initial state, with 72.2% of patients with moderate symptoms remaining moderate. Patients with hypertension had a 67.6% higher risk of progressing from moderate to severe, while those with diabetes had an 89.9% higher risk of deteriorating from severe to critical. Although transition rates to death were low early in hospitalization, these comorbidities significantly increased the likelihood of worsening conditions. Conclusion: This study highlights the utility of continuous-time Markov multistate models in assessing COVID-19 severity progression among hospitalized patients. The findings indicate that patients are more likely to recover than to experience worsening conditions. However, hypertension and diabetes significantly increase the risk of severe outcomes, underscoring the importance of managing these conditions in COVID-19 patients. © The Author(s) 2025.

关键词： Comorbidity COVID-19 progression Markov multistate model Severity state Transition intensity

来源：评论

学校读者我要写书评

暂无评论

Deep learning health space model for ordered responses

引用

BMC Medical Informatics and Decision Making 2025年第1期25卷 1-10页

作者： Lee, Chanhee Park, Taesung Interdisciplinary Program in Bioinformatics Seoul National University Seoul 08826 South Korea Department of Statistics Seoul National University Seoul 08826 South Korea

Background: As personalized medicine becomes more prevalent, the objective measurement and visualization of an individual’s health status are becoming increasingly crucial. However, as the dimensions of data collected from each individual increase, this task becomes more challenging. The Health Space (HS) model provides a statistical framework for visualizing an individual’s health status on biologically meaningful axes. In our previous study, we developed HS models using statistical models such as logistic regression model (LRM) and the proportional odds model (POM). However, these statistical HS models are limited in their ability to accommodate complex non-linear biological relationships. Methods: In order to model complex non-linear biological relationship, we developed deep learning HS models. Specifically, we formulated five distinct deep learning HS models: four standard binary deep neural networks (DNNs) for binary outcomes and one deep ordinal neural network (DONN) that accounts for the ordinality of the dependent variable. We trained these models using 32,140 samples from the Korea National Health and Nutrition Examination Survey (KNHANES) and validated them with data from the Ewha-Boramae cohort (862 samples) and the Korea Association Resource (KARE) project (3,199 samples). Results: The proposed deep learning HS models were compared with the existing statistical HS model based on the POM. Deep learning HS model using DONN demonstrated the best performance in discriminating health status in both the training and external datasets. Conclusion: We developed deep learning HS models to capture complex non-linear biological relationships in HS and compared their performance with our previously best-performing statistical HS model. The deep learning HS models show promise as effective tools for objectively and meaningfully visualizing an individual’s health status. Clinical trial number: Not applicable. © The Author(s) 2025.

关键词： Biologically interpretable visualization Deep ordinal neural network Health space model

来源：评论

学校读者我要写书评

暂无评论

Fate-tox: fragment attention transformer for E(3)-equivariant multi-organ toxicity prediction

引用

Journal of Cheminformatics 2025年第1期17卷 1-16页

作者： Ha, Sumin Bang, Dongmin Kim, Sun Interdisciplinary Program in Artificial Intelligence Seoul National University Seoul 08826 South Korea Interdisciplinary Program in Bioinformatics Seoul National University Seoul 08826 South Korea AIGENDRUG Co. Ltd. Seoul 08758 South Korea Department of Computer Science and Engineering Seoul National University Seoul 08826 South Korea

Toxicity is a critical hurdle in drug development, often causing the late-stage failure of promising compounds. Existing computational prediction models often focus on single-organ toxicity. However, avoiding toxicity of an organ, such as reducing gastrointestinal side effects, may inadvertently lead to toxicity in another organ, as seen in the real case of rofecoxib, which was withdrawn due to increased cardiovascular risks. Thus, simultaneous prediction of multi-organ toxicity is a desirable but challenging task. The main challenges are (1) the variability of substructures that contribute to toxicity of different organs, (2) insufficient power of molecular representations in diverse perspectives, and (3) explainability of prediction results especially in terms of substructures or potential toxicophores. To address these challenges with multiple strategies, we developed FATE-Tox, a novel multi-view deep learning framework for multi-organ toxicity prediction. For variability of substructures, we used three fragmentation methods such as BRICS, Bemis-Murcko scaffolds, and RDKit Functional Groups to formulate fragment-level graphs so that diverse substructures can be used to identify toxicity for different organs. For insufficient power of molecular representations, we used molecular representations in both 2D and 3D perspectives. For explainability, our fragment attention transformer identifies potential 3D toxicophores using attention coefficients. Scientific contribution: Our framework achieved significant improvements in prediction performance, with up to 3.01% gains over prior baseline methods on toxicity benchmark datasets from MoleculeNet (BBBP, SIDER, ClinTox) and TDC (DILI, Skin Reaction, Carcinogens, and hERG), while the multi-task learning approach further enhanced performance by up to 1.44% compared to the single-task learning framework that had already surpassed these baselines. Additionally, attention visualization aligning with literature contributes to

关键词： Graph neural network Multi-task learning Toxicity prediction Transformer

来源：评论

学校读者我要写书评

暂无评论

MV-CLAM: Multi-View Molecular Interpretation with Cross-Modal Projection via Language Model

arXiv

引用

arXiv 2025年

作者： Ha, Sumin Kim, Jun Hyeong Piao, Yinhua Kim, Sun Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of Bio-MAX/N-Bio Seoul National University Korea Republic of Department of Computer Science and Engineering Seoul National University Korea Republic of Interdisciplinary Program in Bioinformatics Seoul National University Korea Republic of AIGENDRUG Co. Ltd. Korea Republic of

Human expertise in chemistry and biomedicine relies on contextual molecular understanding, a capability that large language models (LLMs) can extend through fine-grained alignment between molecular structures and text. Recent multimodal learning advances focus on cross-modal alignment, but existing molecule-text models ignore complementary information in different molecular views and rely on single-view representations, limiting molecular understanding. Moreover, naïve multi-view alignment strategies face two challenges: (1) separate aligned spaces with inconsistent mappings between molecule and text embeddings, and that (2) existing loss objectives fail to preserve complementary information for fine-grained alignment. This can limit the LLM’s ability to fully understand the molecular properties. To address these issues, we propose MV-CLAM, a novel framework that aligns multi-view molecular representations into a unified textual space using a multi-query transformer (MQ-Former). Our approach ensures cross-view consistency while a token-level contrastive loss preserves diverse molecular features across textual queries. MV-CLAM enhances molecular reasoning, improving retrieval and captioning accuracy. The source code of MV-CLAM is available in https://***/sumin124/***. © 2025, CC BY-NC-SA.

关键词： Molecular docking

来源：评论

学校读者我要写书评

暂无评论

Deep learning-based denoising for unbiased analysis of morphology and stiffness in amyloid fibrils

引用

Computers in Biology and Medicine 2025年 184卷 109410-109410页

作者： Park, Jaehee Cheong, Da Yeon Lee, Gyudo Han, Cheol E. Department of Electronics and Information Engineering Korea University Sejong30019 Korea Republic of Interdisciplinary Graduate Program for Artificial Intelligence Smart Convergence Technology Korea University Sejong30019 Korea Republic of Department of Biotechnology and Bioinformatics Korea University Sejong30019 Korea Republic of

Understanding the morphology of amyloid fibrils is crucial for comprehending the aggregation and degradation mechanisms of abnormal proteins implicated in various diseases, such as Alzheimer's disease, Parkinson's disease, type II diabetes, and various forms of amyloidosis. Atomic force microscopy (AFM) stands as the most representative method for studying amyloid fibril morphology. However, obstacles in AFM images, including noise, salt, and amorphous aggregates, often impede accurate sample quantification. In this study, we developed denoising software employing a U-Net deep learning architecture to address this issue. The software efficiently eliminated various impediments that interfere with fibril analysis in noisy AFM images, thereby facilitating precise quantification of amyloid fibrils. We also developed automated fibril analysis technologies using the denoised AFM images, leading to quicker, more precise, and more objective assessments of fibril morphology. Furthermore, we presented a method for fibril stiffness extraction from a modulus image through mask creation based on a denoised height image. Our approach secures time efficiency and precision in analyzing amyloid morphology, and we believe it will significantly advance the currently stagnant research on amyloid-related diseases. © 2024 The Authors

关键词： Neurodegenerative diseases

来源：评论

学校读者我要写书评

暂无评论

AlphaFold Protein Structure Database and 3D-Beacons: New Data and Capabilities

引用

Journal of Molecular Biology 2025年 168967页

作者： Fleming, Jennifer Magana, Paulyna Nair, Sreenath Tsenkov, Maxim Bertoni, Damian Pidruchna, Ivanna Lima Afonso, Marcelo Querino Midlik, Adam Paramval, Urmila Žídek, Augustin Laydon, Agata Kovalevskiy, Oleg Pan, Joshua Cheng, Jun Avsec, Žiga Bycroft, Clare Wong, Lai Hong Last, Meera Mirdita, Milot Steinegger, Martin Kohli, Pushmeet Váradi, Mihály Velankar, Sameer European Molecular Biology Laboratory European Bioinformatics Institute Hinxton United Kingdom Google DeepMind London United Kingdom School of Biological Sciences Seoul National University Seoul 08826 Cuba Interdisciplinary Program in Bioinformatics Seoul National University Seoul 08826 Cuba Institute of Molecular Biology and Genetics Seoul 08826 Cuba Artificial Intelligence Institute Seoul National University Seoul 08826 Cuba

The AlphaFold Protein Structure Database (https://***/) has made significant strides in enhancing its utility and accessibility for the life science research community. The recent integration of AlphaMissense predictions enables access to the pathogenicity of human protein missense variants, with an innovative and interactive heatmap and 3D visualisation that display variant data at the residue level. Users can now toggle between structure model quality (pLDDT) and average pathogenicity scores, providing insights into the implications of specific residue changes. The Foldseek integration offers a rapid and accurate method for protein structure searches and comparisons. Bulk data download options further facilitate comprehensive data analysis and integration with other computational tools. The 3D-Beacons framework (https://***/pdbe/pdbe-kb/3dbeacons/) has also been enhanced with detailed annotation endpoints (such as AlphaMissense data) and integrates LevyLab's dataset of homomeric AlphaFold 2 models. These advancements significantly improve the functionality and accessibility of these resources, enabling discoveries using structure data. © 2025 The Author(s)

关键词： AlphaFold Alphamissense Annotations Foldseek Training

来源：评论

学校读者我要写书评

暂无评论

Pseudomonas aeruginosa-derived DnaJ induces TLR2 expression through TLR10-mediated activation of the PI3K-SGK1 pathway in macrophages

引用

Microbes and Infection 2025年 105481页

作者： Lee, Jaehoo Jin, Yongxin Wu, Weihui Lee, Yeji Ha, Un-Hwan Department of Biotechnology and Bioinformatics Korea University Sejong 30019 South Korea Interdisciplinary Graduate Program for Artificial Intelligence Smart Convergence Technology Korea University Sejong 30019 South Korea State Key Laboratory of Medicinal Chemical Biology Key Laboratory of Molecular Microbiology and Technology of the Ministry of Education Department of Microbiology Nankai University Tianjin 300071 China

TLR2 is a key component of the innate immune system, responsible for recognizing Gram-positive bacterial components and initiating inflammatory signaling cascades that activate defense responses. However, little is known about the regulatory effects of Pseudomonas aeruginosa (P. aeruginosa) on TLR2 expression. In this study, we investigated the potential link between P. aeruginosa-derived DnaJ and TLR2 expression in macrophages, as well as the activation of downstream signaling pathways. Our findings revealed that DnaJ significantly induced TLR2 expression in a dose- and time-dependent manner, predominantly affecting TLR2 with minimal impact on other TLRs, such as TLR4 and TLR5, which detect bacterial PAMPs. The DnaJ-mediated TLR2 induction was driven by activation of the PI3K-SGK1 signaling pathway, with TLR10 playing a crucial role in facilitating these effects. This increase in TLR2 expression led to enhanced production of inflammatory cytokines in response to secondary Staphylococcus aureus infections, indicating a role in boosting host defense mechanisms. In conclusion, these findings suggest that P. aeruginosa-derived DnaJ promotes TLR2 expression via TLR10-mediated activation of the PI3K-SGK1 pathway, thereby enhancing host immune responses against Gram-positive bacterial infections. © 2025 Institut Pasteur

关键词： DnaJ Pseudomonas aeruginosa SGK1 TLR10 TLR2

来源：评论

学校读者我要写书评

暂无评论

CIWARS: A Web Server for Antibiotic Resistance Surveillance Using Longitudinal Metagenomic Data

引用

Journal of Molecular Biology 2025年

作者： Emon, Muhit Islam Cheung, Yat Fei Stoll, James Rumi, Monjura Afrin Brown, Connor Choi, Joung Min Moumi, Nazifa Ahmed Ahmed, Shafayat Song, Haoqiu Sein, Justin Yao, Shunyu Khan, Ahmad Gupta, Suraj Kulkarni, Rutwik Butt, Ali Vikesland, Peter Pruden, Amy Zhang, Liqing Department of Computer Science Virginia Tech Blacksburg 24060 VA United States Fralin Life Science Institute Virginia Tech Blacksburg 24060 VA United States Department of Civil and Environmental Engineering Virginia Tech Blacksburg 24060 VA United States The Interdisciplinary PhD Program in Genetics Bioinformatics and Computational Biology Virginia Tech Blacksburg 24060 VA United States

The rise of antibiotic resistance (AR) poses a substantial threat to human and animal health, food security, and economic stability. Wastewater-based surveillance (WBS) has emerged as a powerful strategy for population-level AR monitoring, providing valuable data to guide public health and policy decisions. Metagenomic sequencing is especially promising, as it can yield comprehensive profiles of antibiotic resistance genes (ARGs) and other genes relevant to AR in a single run. However, online analytical platforms to facilitate analysis of longitudinal metagenomic data are lacking. To address this, we introduce CyberInfrastructure for Waterborne Antibiotic Resistance Surveillance (CIWARS), a web server configured for characterizing key AR trends from longitudinal metagenomic WBS data. CIWARS offers comprehensive profiling of ARGs and taxonomic profiling of pathogen-associated bacterial taxonomic groups, identifies potential associations of ARGs with mobile genetic elements (MGEs) and pathogen-containing taxa, and assesses resistome risk based on the co-occurrence of ARGs, MGEs, and pathogen-like sequences. Additionally, it detects anomalous AR indicators over time, aiding in identifying potential events of concern, such as the emergence of resistant strains or outbreaks. Through interactive temporal data visualization, CIWARS enables AR monitoring and can serve as a tool to inform effective and timely interventions to mitigate the spread and transmission of AR. Here, CIWARS is demonstrated using longitudinal metagenomic data from a wastewater treatment plant (WWTP) influent and effluent, but it can be extended to any environment. CIWARS provides a valuable tool to support global efforts to combat the evolution and spread of AR, while also guiding agricultural and public health efforts aimed at optimizing antibiotic use. The web server is freely available at https://***/. © 2025 Elsevier Ltd

关键词： anomaly detection antibiotic resistance (AR) AR indicator profiling longitudinal metagenomics wastewater-based surveillance

来源：评论

学校读者我要写书评

暂无评论

SAGE-Amine: Generative Amine design with multi-property optimization for efficient CO 2 capture

Carbon Capture Science & Technology

引用

Carbon Capture Science & Technology 2025年 16卷

作者： Hocheol Lim Hyein Cho Jeonghoon Kim Kyoung Tai No Bioinformatics and Molecular Design Research Center (BMDRC) Incheon 21983 Republic of Korea The Interdisciplinary Graduate Program in Integrative Biotechnology & Translational Medicine Yonsei University Incheon 21983 Republic of Korea

Efficient CO 2 capture is vital for mitigating climate change, with amine-based solvents being widely used due to their strong reactivity with CO 2 . However, optimizing key properties such as basicity, viscosity, and absorption capacity remains challenging, as traditional methods rely on labor-intensive experimentation and predefined chemical databases, limiting the exploration of novel solutions. Here, SAGE-Amine was introduced, a generative modeling approach that integrates Scoring-Assisted Generative Exploration (SAGE) with quantitative structure-property relationship models to design new amines tailored for CO 2 capture. Unlike conventional virtual screening restricted to existing compounds, SAGE-Amine generates novel amines by leveraging autoregressive natural language processing models trained on amine datasets. SAGE-Amine identified known amines for CO 2 capture from scratch and successfully performed single-property optimization, increasing basicity or reducing viscosity or vapor pressure. Furthermore, it facilitated multi-property optimization, simultaneously achieving high basicity with low viscosity and vapor pressure. The 10 top-ranked amines were suggested using SAGE-Amine and their thermodynamic properties were further assessed using COSMO-RS simulations, confirming their potential for CO 2 capture. These results highlight the potential of generative modeling in accelerating the discovery of amine solvents and expanding the possibilities for industrial CO 2 capture applications.

关键词： Material discovery De novo molecular design Quantitative structure-property relationship Solvent design Amine design

来源：评论

学校读者我要写书评

暂无评论

Cholesterol allosteric modulation of the oxytocin receptor

引用

Biophysical Journal 2025年

作者： Marlow, Brennica Vogel, Alexander Kuenze, Georg Pankonin, Maik Reinhardt, Franziska Stadler, Peter F. Hildebrand, Peter W. Meiler, Jens Center for Structural Biology Vanderbilt University Nashville TN United States Institute for Drug Discovery Medical Faculty Leipzig University Leipzig Germany Center for Scalable Data Analytics and Artificial Intelligence ScaDS.AI Leipzig University Leipzig Germany Interdisciplinary Center for Bioinformatics Leipzig University Leipzig Germany Chemical and Physical Biology Program Vanderbilt University Nashville TN United States Department of Chemistry Vanderbilt University Nashville TN United States Institute of Medical Physics and Biophysics Medical Faculty Leipzig University Leipzig Germany Bioinformatics Group Institute of Computer Science Interdisciplinary Center of Bioinformatics Leipzig University Leipzig Germany Max-Planck-Institute for Mathematics in the Sciences Leipzig Germany Institute of Theoretical Chemistry University of Vienna Wien Austria Facultad de Ciencias Universidad National de Colombia Sede Bogota Colombia Santa Fe Institute New Mexico Santa Fe Mexico

G-protein coupled receptors are critical components in cellular signaling, mediating various physiological responses to external stimuli. Here, we investigate the intricate relationship between cholesterol and the oxytocin receptor (OXTR), focusing on the binding mechanisms and the allosteric cross talk of bound cholesterol to the orthosteric ligand binding pocket. Utilizing molecular docking and molecular dynamics simulations, we identify cholesterol binding sites both on the agonist-bound and antagonist-bound state, which show differing distributions and residence times of the cholesterol molecules. Importantly, both methods converge on several key sites, demonstrating strong predictive overlap. Notably, one such site, and several sites detected by our molecular dynamics approach, also coincide with electron density observed in an experimental cryo-EM map, providing orthogonal validation for computational predictions. Allosteric network analysis uncovers the distinct pathways through which cholesterol may affect ligand-mediated receptor signaling, highlighting the significance of one site on the extracellular leaflet between TM4 and TM5, and two sites on the intracellular leaflet between TM2, TM3, and TM4 and between TM4 and TM5 in transmitting allosteric signals to the orthosteric pocket. These findings provide insights into the impact of cholesterol on OXTR function, emphasizing specific binding sites and signaling paths for further experimental exploration. © 2025 Biophysical Society

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：