面向数据驱动研究的佐剂数据库与知识库：计算生物学视角

摘要(中文译文)

在佐剂研究中，长期以来依赖经验和反复试验，而包括人工智能（AI）/机器学习在内的广泛领域的技术进步推动了数据驱动研究的发展，支持了更多基于证据的高效方法的引入。数据驱动研究，包括疫苗学中的系统生物学（“系统疫苗学”），不仅需要分析技术或方法，还需要数据库和知识库等资源。本文介绍了目前可用的佐剂相关数据库，包括可用于数据驱动研究的数据集和知识库。通过将其分为定量和定性数据，介绍了若干不同数据驱动方法的实例，如系统疫苗学、用于知识图谱的图神经网络、基于大语言模型的AI智能体，说明每种类型数据如何被利用，以表明数据库和知识库维护与管理的重要性。

Abstract

In adjuvant research, which has long relied on experience and trial and error, advancements in technologies in broad areas, including artificial intelligence (AI)/machine learning, enabling data-driven research, have supported the introduction of more evidence-based, efficient approaches. Data-driven research, including systems biology in vaccinology ('systems vaccinology'), requires not only analytical technologies or techniques but also resources such as databases and knowledge bases. This paper introduces currently available adjuvant-related databases, including datasets utilizable for data-driven research and knowledge bases. By categorizing them into quantitative and qualitative data, several examples of different data-driven approaches, such as systems vaccinology, graph neural network for knowledge graph, AI agent based on a large language model, illustrating how each type of data is utilized, are introduced to show the importance of maintenance and management of the databases and knowledge bases.

如何引用

Current opinion in virology 的其他文献

• 产碱菌脂质A：一种独特的TLR4激动剂粘膜佐剂，诱导分泌型IgA和Th17反应。