Under the category of traditional Chinese medicine's analogical thinking, the protein-protein interaction network of "treating skin with skin", the pathway map of enrichment analysis, and the complex network diagram of "core drug-component-target-pathway"

The intersection of targets corresponding to 15 high-frequency drugs and high-frequency traditional Chinese medicines (TCMs) with disease-specific targets yielded a total of 397 targets. Among them, the intersection of targets from six core bark TCMs and disease targets yielded 212 "drug-disease" intersecting targets, accounting for 53.4% of the total intersecting targets between high-frequency drugs, high-frequency TCMs, and diseases. This fully demonstrates the significant role of the six core bark TCMs in the treatment of Alzheimer's Disease (AD). The intersecting target genes selected from the intersection of AD and core prescription drug components were imported into the String platform, with the species selected as Homo sapiens. This resulted in the construction of a protein-protein interaction (PPI) network. Using the CentiScaPe 2.2 component of Cytoscape 3.10.1 software, relevant parameters of the intersecting targets were calculated. Core targets were selected based on the criteria of degree ≥26.437, betweenness centrality ≥247.136, and closeness centrality ≥0.002 (all values rounded to three decimal places). The 212 intersecting targets were analyzed using the String platform, with the species selected as Homo sapiens, to obtain the PPI network diagram. The network analysis results showed that the network, under the system's default medium confidence level (0.4), contained 212 nodes (representing protein molecules) and 2,723 edges (representing interactions between proteins), with an average node degree of 25.9 and an average local clustering coefficient of 0.554. In the graph, nodes represent proteins, and edges represent relationships between proteins. The more edges between two nodes, the darker the color, indicating a greater correlation between their interactions. Subsequently, a biological pathway enrichment analysis was conducted, and the most representative human (hsa) biological species metabolic pathways among the enriched pathway results were analyzed based on their P values, Count values, etc. Finally, using Cytoscape 3.10.1 software, a "core drug-component-target-pathway" complex network diagram was generated to complete the visual analysis. Further analysis of the KEGG-hsa05200-organism-specific pathway (ORG) anchors gene and target information to specific biological functional pathways, facilitating the exploration of the association between core functions and diseases and enabling further interpretation from "gene list" to "functional mechanism". The analysis results of KEGG-hsa05200 (Figure 12) indicate that there is a significant relationship between multiple genes and positive regulation of phosphorylation between the cytoplasm and the nucleus, which is closely related to the cAMP signaling pathway, PI3K-Akt signaling pathway, and MAPK signaling pathway. The downstream processes within the nucleus are mostly indirectly regulating various processes such as cell cycle, apoptosis, and autophagy. The KEGG-hsa05200-ORG diagram integrates core signaling axes such as PI3K-Akt, MAPK, NF-κB, and JAK-STAT. These pathways are not only the main downstream transmission pathways of IL-6, TNF, and IL-1β, but also the core regulatory network for keratinocyte activation, immune cell infiltration, and inflammation persistence in AD. IL-6 regulates immune imbalance in AD through the hsa05200 core branch JAK2-STAT3 pathway; TNF-α activates the NF-κβ and MAPK pathways in hsa05200 by binding to Tumor Necrosis Factor Receptor 1/2 (TNFR1/2), upregulating Matrix Metalloproteinases (MMPs) and Cyclooxygenase-2 (COX-2), which can amplify the dominant immune response in AD, disrupt the skin barrier, and promote the release of inflammatory mediators; IL-1β, as a core immune factor, activates the NF-κβ and p38-MAPK signaling pathways in hsa05200 through the MyD88-dependent pathway, inducing keratinocytes to secrete secondary inflammatory factors such as IL-6. The disease name of AD, six high-frequency core traditional Chinese medicine (TCM) dermatological drugs, and their corresponding 46 main active ingredients, 781 gene targets, top 20 enriched analysis pathways, and core targets on each pathway were imported into Cytoscape 3.10.1 to construct a biological information network, as shown in Figure 13. In this network diagram, the node degree value represents the number of lines connected to the node in the network. The higher the degree value, the higher the correlation and the more prominent the key role. The red circular target in the figure represents the disease AD, the yellow diamond target represents the core composition of dermatological drugs, the peripheral green irregular target represents the main active ingredients of the drugs, and the blue square target represents the target genes. The results showed that the network contained a total of 854 nodes and 1,911 edges, with a network centralization index of 0.182 and a characteristic path length of 3.985. In the visualization results, the red circular targets represent the disease AD, the yellow diamond targets represent the core skin drug components, the peripheral green irregular targets represent the main active ingredients of the drugs, and the blue square targets represent the target genes. The relationships between nodes are closely connected. The node degree value represents the number of lines connected to the node in the network. The larger the degree value, the higher the correlation and the more prominent the key role. According to the CentiScaPe 2.2 platform-assisted calculation, the degree threshold of this network is 4.475, the betweenness centrality threshold is 2,546.473, and the closeness centrality threshold is 2.975. The top three degree values are all drug components, namely vitamin E (159), vitamin B (134), and trigonelline (111).

将15种高频化学药物与高频中药材（traditional Chinese medicines, TCMs）的对应靶点与疾病特异性靶点取交集，共获得397个靶点。其中，6种核心皮类中药材的靶点与疾病靶点的交集得到212个“药物-疾病”交集靶点，占高频化学药物、高频中药材与疾病三者交集靶点总数的53.4%，这充分证明了该6种核心皮类中药材在阿尔茨海默病（Alzheimer's Disease, AD）治疗中的重要作用。将从阿尔茨海默病与核心处方药物成分的交集筛选得到的交集靶基因导入String平台，选择物种为智人（Homo sapiens），构建蛋白质相互作用（protein-protein interaction, PPI）网络。使用Cytoscape 3.10.1软件的CentiScaPe 2.2组件，计算交集靶点的相关参数，基于度（degree）≥26.437、介数中心性（betweenness centrality）≥247.136、紧密度中心性（closeness centrality）≥0.002（所有数值均保留三位小数）的标准筛选核心靶点。对212个交集靶点通过String平台进行分析，选择物种为智人，获得PPI网络图。网络分析结果显示，在系统默认的中等置信度（0.4）下，该网络包含212个节点（代表蛋白质分子）与2723条边（代表蛋白质间相互作用），平均节点度为25.9，平均局部聚类系数为0.554。网络图中，节点代表蛋白质，边代表蛋白质间的相互关系，两个节点间的边数越多，颜色越深，表明二者相互作用的相关性越强。随后进行生物通路富集分析，基于P值、Count值等指标，对富集通路结果中最具代表性的人类（hsa）生物物种代谢通路进行分析。最终使用Cytoscape 3.10.1软件生成“核心药物-成分-靶点-通路”复合网络图，完成可视化分析。进一步对KEGG-hsa05200-物种特异性通路（organism-specific pathway, ORG）进行分析，将基因与靶点信息锚定至特定生物功能通路，便于探索核心功能与疾病的关联，实现从“基因列表”到“功能机制”的进一步阐释。KEGG-hsa05200分析结果（图12）显示，多个基因与细胞质至细胞核的磷酸化正调控存在显著关联，这与cAMP信号通路、PI3K-Akt信号通路及MAPK信号通路密切相关。细胞核内的下游过程多间接调控细胞周期、细胞凋亡、细胞自噬等多种进程。KEGG-hsa05200-ORG图整合了PI3K-Akt、MAPK、NF-κB及JAK-STAT等核心信号轴，这些通路不仅是IL-6、TNF及IL-1β的主要下游传导通路，同时也是阿尔茨海默病中角质形成细胞活化、免疫细胞浸润及炎症持续的核心调控网络。IL-6通过hsa05200核心分支JAK2-STAT3通路调控阿尔茨海默病的免疫失衡；TNF-α通过结合肿瘤坏死因子受体1/2（Tumor Necrosis Factor Receptor 1/2, TNFR1/2）激活hsa05200中的NF-κβ及MAPK通路，上调基质金属蛋白酶（Matrix Metalloproteinases, MMPs）及环氧合酶-2（Cyclooxygenase-2, COX-2），可放大阿尔茨海默病中的优势免疫反应、破坏皮肤屏障并促进炎症介质释放；IL-1β作为核心免疫因子，通过MyD88依赖通路激活hsa05200中的NF-κβ及p38-MAPK信号通路，诱导角质形成细胞分泌IL-6等继发性炎症因子。将阿尔茨海默病（AD）、6种高频核心皮肤科中药材、其对应的46种主要活性成分、781个基因靶点、富集分析排名前20的通路，以及各通路上的核心靶点导入Cytoscape 3.10.1软件，构建生物信息网络，如图13所示。该网络图中，节点度值代表网络中与该节点相连的边数，度值越高，相关性越强，关键作用越突出。图中红色圆形节点代表疾病阿尔茨海默病，黄色菱形节点代表核心皮肤科药物成分，外围绿色不规则节点代表药物主要活性成分，蓝色方形节点代表靶基因。结果显示，该网络共包含854个节点与1911条边，网络中心化指数为0.182，特征路径长度为3.985。可视化结果中，红色圆形节点代表疾病阿尔茨海默病，黄色菱形节点代表核心皮肤药物成分，外围绿色不规则节点代表药物主要活性成分，蓝色方形节点代表靶基因，节点间关联紧密。节点度值代表网络中与该节点相连的边数，度值越大，相关性越高，关键作用越显著。经CentiScaPe 2.2平台辅助计算，该网络的度阈值为4.475，介数中心性阈值为2546.473，紧密度中心性阈值为2.975。度值排名前三的均为药物成分，分别为维生素E（159）、维生素B（134）及葫芦巴碱（trigonelline, 111）。