基于基因组数据分析的细菌耐药基因识别与表型预测

Identification and prediction of bacterial antibiotic resistance via genomic data analysis

摘要:

利用基因组数据和生物信息学分析方法，快速鉴定耐药基因并预测耐药表型，为细菌耐药状况监测提供了有力辅助手段。目前，已有的数十个耐药数据库及其相关分析工具这些资源为细菌耐药基因的识别以及耐药表型的预测提供了数据信息和技术手段。随着细菌基因组数据的持续增加以及耐药表型数据的不断积累，大数据和机器学习能够更好地建立耐药表型与基因组信息之间的相关性，因此，构建高效的耐药表型预测模型成为研究热点。本文围绕细菌耐药基因的识别和耐药表型的预测，针对耐药相关数据库、耐药特征识别理论与方法、耐药数据的机器学习与表型预测等方面展开讨论，以期为细菌耐药的相关研究提供手段和思路。

Abstract:

Using genomic data and bioinformatics methods has become an important approach to rapidly identify the genes and predict the phenotypes of bacterial antibiotic resistance. Dozens of antibiotic resistance databases have been established, providing information and auxiliary tools for the identification and prediction of bacterial antibiotic resistance. As the bacterial genome data and antibiotic resistance phenotype data are increasing, the correlation between them can be established via big data and machine learning. Therefore, establishing efficient models predicting antibiotic resistance phenotypes has become a research hot topic. Focusing on the gene identification and phenotype prediction of bacterial antibiotic resistance, this review discusses the related databases, the theories and methods, the machine learning algorithms, and the prediction models. In addition, we made an outlook on the future prospects in this field, aiming to provide new ideas for the related studies.