Viral Sentry AI-Automated zoonotic surveillance and drug repurposing agent.

IF 2.4 Q3 BIOCHEMICAL RESEARCH METHODS
Biology Methods and Protocols Pub Date : 2026-05-27 eCollection Date: 2026-01-01 DOI:10.1093/biomethods/bpag026
Cristian R Munteanu, Jose Vázquez-Naya, Eduardo Tejera
{"title":"Viral Sentry AI-Automated zoonotic surveillance and drug repurposing agent.","authors":"Cristian R Munteanu, Jose Vázquez-Naya, Eduardo Tejera","doi":"10.1093/biomethods/bpag026","DOIUrl":null,"url":null,"abstract":"<p><p>Zoonotic viruses capable of jumping from animal reservoirs into human populations represent a persistent and unpredictable menace to global health. To confront this challenge, we developed Viral Sentry AI, an autonomous agent designed to close the gap between viral emergence and therapeutic response. Unlike static analysis tools, Viral Sentry AI operates as a continuous sentinel, automatically scanning the National Center for Biotechnology Information public databases for new viral genomes and executing a three-stage agentic surveillance workflow, with distinct, specialized artificial intelligence architectures for generated text, macromolecule sequences, and drug chemical data. First, the system is using a Large Language Model (Gemma4) to parse unstructured submission records and extract the host information if it is not available in the dedicated field. In the second stage, the system employs a novel deep-learning topology, virsentai-v3-hyena-dna-16k, a fine-tuned HyenaDNA model capable of processing complete viral genomes up to 160 000 bases. This architecture captures subtle, long-range genomic dependencies to predict human infectivity with high precision. Upon predicting the possible human infection of the scanned viruses, the agent autonomously triggers a downstream therapeutic module as the stage three. It extracts National Center for Biotechnology Information RefSeq viral protein sequences and utilizes a pretrained Protein-Ligand Affinity Prediction Transformer model to calculate affinity interactions against 2092 ChEMBL-approved drugs, instantly identifying candidates for drug repurposing. In rigorous cross-validation on a curated dataset of 33 426 complete viral genomes, the surveillance module demonstrated robust discriminatory power, achieving an Area Under the Receiver Operating Characteristic Curve of 0.88 in classifying human host potential. By integrating state-of-the-art genomic modeling with automated lead compound screening, Viral Sentry AI offers a proactive, end-to-end research prototype for pandemic preparedness. The platform is freely accessible at https://muntisa.github.io/virsentai (source code: https://github.com/muntisa/virsentai).</p>","PeriodicalId":36528,"journal":{"name":"Biology Methods and Protocols","volume":"11 1","pages":"bpag026"},"PeriodicalIF":2.4000,"publicationDate":"2026-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13256000/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Biology Methods and Protocols","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1093/biomethods/bpag026","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2026/1/1 0:00:00","PubModel":"eCollection","JCR":"Q3","JCRName":"BIOCHEMICAL RESEARCH METHODS","Score":null,"Total":0}
引用次数: 0

Abstract

Zoonotic viruses capable of jumping from animal reservoirs into human populations represent a persistent and unpredictable menace to global health. To confront this challenge, we developed Viral Sentry AI, an autonomous agent designed to close the gap between viral emergence and therapeutic response. Unlike static analysis tools, Viral Sentry AI operates as a continuous sentinel, automatically scanning the National Center for Biotechnology Information public databases for new viral genomes and executing a three-stage agentic surveillance workflow, with distinct, specialized artificial intelligence architectures for generated text, macromolecule sequences, and drug chemical data. First, the system is using a Large Language Model (Gemma4) to parse unstructured submission records and extract the host information if it is not available in the dedicated field. In the second stage, the system employs a novel deep-learning topology, virsentai-v3-hyena-dna-16k, a fine-tuned HyenaDNA model capable of processing complete viral genomes up to 160 000 bases. This architecture captures subtle, long-range genomic dependencies to predict human infectivity with high precision. Upon predicting the possible human infection of the scanned viruses, the agent autonomously triggers a downstream therapeutic module as the stage three. It extracts National Center for Biotechnology Information RefSeq viral protein sequences and utilizes a pretrained Protein-Ligand Affinity Prediction Transformer model to calculate affinity interactions against 2092 ChEMBL-approved drugs, instantly identifying candidates for drug repurposing. In rigorous cross-validation on a curated dataset of 33 426 complete viral genomes, the surveillance module demonstrated robust discriminatory power, achieving an Area Under the Receiver Operating Characteristic Curve of 0.88 in classifying human host potential. By integrating state-of-the-art genomic modeling with automated lead compound screening, Viral Sentry AI offers a proactive, end-to-end research prototype for pandemic preparedness. The platform is freely accessible at https://muntisa.github.io/virsentai (source code: https://github.com/muntisa/virsentai).

病毒哨兵ai -自动人畜共患病监测和药物再利用剂。
能够从动物宿主跳入人群的人畜共患病毒对全球健康构成持续和不可预测的威胁。为了应对这一挑战,我们开发了病毒哨兵人工智能,这是一种自主代理,旨在缩小病毒出现和治疗反应之间的差距。与静态分析工具不同,病毒哨兵人工智能作为一个连续的哨兵,自动扫描国家生物技术信息中心的公共数据库,寻找新的病毒基因组,并执行一个三阶段的代理监测工作流程,使用独特的、专门的人工智能架构来生成文本、大分子序列和药物化学数据。首先,系统使用大型语言模型(Gemma4)来解析非结构化提交记录,并在专用字段中不可用时提取主机信息。在第二阶段,该系统采用了一种新的深度学习拓扑,virsentai-v3-hyena-dna-16k,这是一种经过微调的HyenaDNA模型,能够处理多达16万个碱基的完整病毒基因组。这种结构捕捉到微妙的、长期的基因组依赖性,从而高精度地预测人类的传染性。在预测可能的人类感染扫描病毒后,药剂自动触发下游治疗模块作为第三阶段。它提取国家生物技术信息中心RefSeq病毒蛋白序列,并利用预训练的蛋白质-配体亲和预测转换器模型计算2092种经chembll批准的药物的亲和相互作用,立即确定药物再利用的候选药物。在一个包含33426个完整病毒基因组的数据集上进行了严格的交叉验证,监测模块显示出强大的区分能力,在分类人类宿主潜力方面实现了0.88的接受者工作特征曲线下面积。通过将最先进的基因组建模与自动先导化合物筛选相结合,Viral Sentry AI为大流行防范提供了主动的端到端研究原型。该平台可在https://muntisa.github.io/virsentai免费访问(源代码:https://github.com/muntisa/virsentai)。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Biology Methods and Protocols
Biology Methods and Protocols Agricultural and Biological Sciences-Agricultural and Biological Sciences (all)
CiteScore
3.80
自引率
2.80%
发文量
28
审稿时长
19 weeks
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信
小红书