摘要
目的 肝癌，作为一种全球性的严重健康威胁，其发病机制错综复杂，而传统治疗方法效果有限。生物信息学和机器学习技术的融合，为在基因数据库中识别与疾病相关的基因提供了可能，为开发创新治疗方法和确定新的靶点提供了潜在途径。方法 下载人肝癌和正常的单细胞转录组测序数据，通过R程序整理分析找到在正常和疾病组差异较大的基因，通过多种机器学习进行筛选获得核心基因，然后验证核心基因的预后效果并检验其与免疫细胞和功能的相关性。结果 我们发现一批显著变化的基因。借助机器学习分析，提炼并确认了11个核心基因。同时，我们预测了核心基因的表达水平与患者的生存时间的关联揭示了核心基因与免疫细胞之间的紧密关系。结论 我们的研究在单细胞水平上展示了人的肝细胞癌模型，揭示了癌变肝脏的关键特征，包括T细胞 NK细胞的增多并通过机器学习挑选核心基因并证实了LY6E可能是影响肝癌进展的关键基因。
关键词： 肝细胞癌；机器学习；单细胞测序
Abstract
Objective Liver cancer, a serious global health threat, presents complex pathogenesis and limited efficacy of traditional treatments. The integration of bioinformatics and machine learning offers a pathway to identify disease-related genes in genetic databases, paving the way for developing innovative treatments and identifying new targets. Methods Human liver cancer and normal single-cell transcriptome sequencing data were downloaded and analyzed using R programming to identify genes with significant differences between normal and disease groups. Various machine learning methods were employed to screen for core genes, which were then validated for their prognostic significance and examined for their association with immune cells and functions. Results A set of significantly altered genes was identified. Through machine learning analysis, 11 core genes were refined and confirmed. Additionally, the association between the expression levels of core genes and patient survival time was predicted, revealing a close relationship between these genes and immune cells. Conclusion Our study presents a human liver cancer model at the single-cell level, revealing key characteristics of cancerous liver, including the increase of T cells and NK cells. Machine learning was utilized to select core genes, confirming that LY6E may be a critical gene influencing the progression of liver cancer.
Key words： Hepatocellular carcinoma; Machine learning; Single cell sequencing
参考文献 References
[1] Kanwal F, Singal AG. Surveillance for Hepatocellular Carcinoma: Current Best Practice and Future Direction. [J]. Gastroenterology 2019; 157(1): 54-64 [PMID: 30986389 PMCID: Q1.
[2] Hepatocellular carcinoma. [J].Nat Rev Dis Primers 2021; 7(1): 7 [PMID: 33479233 PMCID: Q1.
[3] Morgan RL, Baack B, Smith BD, et al. Eradication of hepatitis C virus infection and the development of hepatocellular carcinoma: a meta-analysis of observational studies. [J].Ann Intern Med 2013; 158(5 Pt 1): 329-337 [PMID: 23460056 PMCID: Q1.
[4] Chen C-J, Yang H-I, Su J, et al. Risk of hepatocellular carcinoma across a biological gradient of serum hepatitis B virus DNA level. [J]. JAMA 2006; 295(1): 65-73 [PMID: 16391218 PMCID: Q1].
[5] Chen J-D, Yang H-I, Iloeje UH, et al. Carriers of inactive hepatitis B virus are still at risk for hepatocellular carcinoma and liver-related death. [J].Gastroenterology 2010; 138(5): 1747-1754 [PMID: 20114048 PMCID: Q1.
[6] Potter SS. Single-cell RNA sequencing for the study of development, physiology and disease. [J].Nature reviews Nephrology 2018; 14(8): 479-492 [PMID: 29789704 PMCID: PMC6070143.
[7] Single-cell transcriptomics of 20 mouse organs creates a Tabula Muris. [J].Nature 2018; 562(7727): 367-372 [PMID: 30283141 PMCID: PMC6642641.
[8] Azuaje F. Artificial intelligence for precision oncology: beyond patient stratification. [J].NPJ precision oncology 2019; 3: 6 [PMID: 30820462 PMCID: PMC6389974.
[9] Barrett T, Wilhite SE, Ledoux P, et al. NCBI GEO: archive for functional genomics data sets--update. [J].Nucleic acids research 2013; 41(Database issue): D991-995 [PMID: 23193258 PMCID: PMC3531084.
[10] Stuart T, Butler A, Hoffman P, et al. Comprehensive Integration of Single-Cell Data. [J].Cell 2019; 177(7): 1888-1902.e1821 [PMID: 31178118 PMCID: PMC 6687398.
[11] Korsunsky I, Millard N, Fan J, et al. Fast, sensitive and accurate integration of single-cell data with Harmony. [J]. Nature methods 2019; 16(12): 1289-1296 [PMID: 31740819 PMCID: PMC6884693.
[12] Aran D, Looney AP, Liu L, et al. Reference-based analysis of lung single-cell sequencing reveals a transitional profibrotic macrophage. [J].Nature immunology 2019; 20(2): 163-172 [PMID: 30643263 PMCID: PMC6340744.
[13] Zhang X, Lan Y, Xu J, et al.CellMarker: a manually curated resource of cell markers in human and mouse. [J]. Nucleic acids research 2019; 47(D1): D721-d728 [PMID: 30289549 PMCID: PMC6323899.
[14] Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. [J].BMC bioinformatics 2013; 14: 7 [PMID: 23323831 PMCID: PMC3618321.
[15] Purdy AK, Campbell KS. Natural killer cells and cancer: regulation by the killer cell Ig-like receptors (KIR). [J].Cancer Biol Ther 2009; 8(23): 2211-2220 [PMID: 19923897 PMCID: PMC2825280.
[16] Song B, Shen S, Fu S, et al. HSPA6 and its role in cancers and other diseases. [J].Mol Biol Rep 2022; 49(11): 10565-10577 [PMID: 35666422.
[17] Huang H, Qiu Y, Huang G, et al. Value of Ferritin Heavy Chain (FTH1) Expression in Diagnosis and Prognosis of Renal Cell Carcinoma. [J].Med Sci Monit 2019; 25: 3700-3715 [PMID: 31104064 PMCID: PMC6537665.
[18] Waring RH. Cytochrome P450: genotype to phenotype. [J].Xenobiotica 2020; 50(1): 9-18 [PMID: 31411087.
[19] Mantovani F, Collavin L, Del Sal G. Mutant p53 as a guardian of the cancer cell. [J]. Cell Death Differ 2019; 26(2): 199-212 [PMID: 30538286 PMCID: PMC6329812.
[20] Pope ED, 3rd, Kimbrough EO, et al. Aberrant lipid metabolism as a therapeutic target in liver cancer. [J]. Expert Opin Ther Targets 2019; 23(6): 473-483 [PMID: 31076001 PMCID: PMC6594827.
[21] Zhou S. Sparse SVM for Sufficient Data Reduction. [J]. IEEE transactions on pattern analysis and machine intelligence 2022; 44(9): 5560-5571 [PMID: 33891547.