AUTHOR=Chen Zeshan , Peng Peichun , Wang Miaodong , Deng Xin , Chen Rudi TITLE=Bioinformatics-based and multiscale convolutional neural network screening of herbal medicines for improving the prognosis of liver cancer: a novel approach JOURNAL=Frontiers in Medicine VOLUME=Volume 10 - 2023 YEAR=2023 URL=https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2023.1218496 DOI=10.3389/fmed.2023.1218496 ISSN=2296-858X ABSTRACT=BACKGROUND: Liver cancer is one of the major diseases threatening human life and health, and the purpose of this work is to explore new methods for treating liver cancer. METHODS: A deep learning model for the efficacy of clinical herbal medicines for liver can cer was constructed based on NDCN, combined with natural evolutionary rules of genetic alg orithm to obtain the herbal compound for liver cancer treatment. We obtained the differential genes between liver cancer tissues and normal tissues from TCGA database analysis, screened the active ingredients and corresponding targets of herbal compound using TCMSP database, mapped the intersection to obtain the potential targets of herbal compound for liver cancer tr eatment in Venny platform, constructed PPI network, and performed GO analysis and KEGG analysis on the targets of herbal compound for liver cancer treatment. Finally, the key active i ngredients and important targets were molecularly docked. Results: ①The accuracy of NDC N training set is 0.92, and the accuracy of test set is 0.84. After combining with genetic algori thm for 1000 iterations, a set of Chinese herbal compound prescriptions is finally output. ②8 6 targets of herbal compound for liver cancer were obtained, mainly five core targets of IL-6, ESR1, JUN, IL1β and MMP9. Among them, quercetin, kaempferol and Stigmasterol may be t he key active ingredients in hepatocellular carcinoma, and the herbal compound may particip ate in inflammatory response and immune regulation process by mediating IL-17 signaling pa thway and TNF signaling pathway, etc. The anti-cancer effects of the herbal compound may b e mediated by IL-17 signaling pathway, TNF signaling pathway and other signaling pathways, which are involved in inflammatory response and immune regulation. Molecular docking sho wed that the three core target proteins produced stable binding to the two main active ingredi ents. CONCLUSION: The screening of effective herbal compound for clinical treatment of li ver cancer based on NDCNN and genetic algorithm is a feasible approach and will provide id eas for the development of herbal medicines for the treatment of liver cancer as well as other cancers.