目的 综合分析老年流感单细胞转录组测序（single cell transcriptome sequencing，scRNA-seq）和转录组测序（transcriptome sequencing，RNA-seq）数据集，构建老年重症流感风险预测模型，并预测调控相关免疫细胞群的中药。通过统计分析中药的性味归经、药物分类，为中医药治疗老年流感提供理论支持。方法 通过scRNA-seq和RNA-seq数据集得到具有较高预测能力的老年流感相关免疫细胞群及其关键基因。基因本体生物学过程（gene ontology biological process，GOBP）、标志性基因集（Hallmark gene set）、京都基因与基因组百科全书（Kyoto encyclopedia of genes and genomes，KEGG）和Reactome生物通路知识库富集分析揭示关键基因参与的生物过程及通路。循环算法用于构建老年重症流感风险预测模型。CTD和COREMINE数据库查找对关键基因有调控作用的化学成分及中药，并通过古今医案云平台统计高频中药的性味归经、药物分类。结果 单核细胞及其相关基因对于老年流感具有较高的预测能力。老年流感单核细胞相关基因显著富集在感染、代谢和炎症免疫等生物过程及信号通路。5个基因（BCL6、PLSCR1、FCGR1A、RETN、RPL9）的组合可较为准确预测老年重症流感。根据参考文献筛选出42种化学成分，得到高频中药石斛、灵芝、地骨皮、人参、桑叶等；药性以平、寒、温为主；药味以甘、苦、辛为主；归经以肝、脾、肺经居多；以补虚药、清热药、利水渗湿药、解表药、活血化瘀药、消食药、止咳平喘药为主。结论 单核细胞可能在老年流感中发挥重要作用。BCL6、PLSCR1、FCGR1A、RETN、RPL9可能是老年重症流感的潜在诊断及治疗靶标。通过生物信息学分析得出老年流感单核细胞相关基因及对应的高频中药，为治疗老年流感的中药新药开发及临床应用提供了新思路。

Objective To comprehensively analyze the single cell transcriptome sequencing (scRNA-seq) and transcriptome sequencing (RNA-seq) data of influenza in the elderly, construct a risk prediction model of severe influenza in the elderly, and predict the Chinese herbal medicines that regulate the relevant immune cell groups. To provide theoretical support for the treatment of influenza in the elderly by statistical analysis of the nature, taste, meridian tropism, and drug classification of traditional Chinese medicine. Methods Influenza-associated immune cell populations and their key genes with high predictive power were obtained from scRNA-seq and RNA-seq data. The enrichment analysis of gene ontology biological process (GOBP), Hallmark gene set, Kyoto encyclopedia of genes and genomes (KEGG) and Reactome revealed the biological processes and pathways involved in key genes. The circulation algorithm was used to construct the risk prediction model of severe influenza in the elderly. CTD and COREMINE databases were used to search for chemical components and traditional Chinese medicines that have regulatory effects on key genes, and the nature, taste, channel tropism, and drug classification of high-frequency traditional Chinese medicines were counted through the cloud platform of ancient and modern medical records. Results Monocytes and their related genes have a high predictive ability for influenza in the elderly. The monocyte-related genes of influenza in the elderly were significantly enriched in biological processes and signal pathways such as metabolism, inflammation, and immunity. The combination of five genes (BCL6, PLSCR1, FCGR1A, RETN, RPL9) can accurately predict severe influenza in the elderly. A total of 42 chemical components were screened according to the references, and the high-frequency traditional Chinese medicines were Shihu (Dendrobii Caulis), Lingzhi (Ganoderma), Digupi (Lycii Cortex), Renshen (Ginseng Radix et Rhizoma), Sangye (Mori Folium), etc. The medicinal properties were mainly flat, cold, and warm; The flavors were mainly sweet, bitter, and pungent; Liver, spleen, and lung channels were the main channels. The main drugs were tonifying deficiency, clearing heat, removing dampness, relieving exterior, promoting blood circulation and removing blood stasis, eliminating food, and relieving cough and asthma. Conclusion Monocytes may play an important role in influenza in the elderly. BCL6, PLSCR1, FCGR1A, RETN and RPL9 may be potential diagnostic and therapeutic targets for severe influenza in the elderly. Bioinformatics analysis has identified the monocyte-related genes and corresponding high-frequency traditional Chinese medicines for influenza in the elderly, providing new insights for the development and clinical application of novel traditional Chinese medicines in the treatment of influenza in the elderly.

Q811.4；R285

国家中医药管理局中医药创新团队及人才支持计划项目（ZYYCXTD-D-202203）；广州市科技局市校院联合资助项目：广州市中西医结合防治新发突发传染病重点实验室（202201020382）；广东省中医药局科研项目（20222071）；广东省中医院重点实验室专项（JZ2020ZZ01，YN2023JZ02）；广东省中医院朝阳人才科研专项资助（ZY2022KY10，ZY2022YL04）