Establishment of multifactor predictive models for the occurrence and progression of cervical intraepithelial neoplasia

BMC Cancer

Mengjie Chen¹  He Wang¹  Li Li¹  Mingmiao Hu¹  Yuejuan Liang¹ 

[1] Guangxi Medical University affiliated Cancer Hospital, NO.71 Hedi Road Qingxiu Square, Nanning City, Guangxi Province, China;

关键词: Cervical intraepithelial neoplasia;    Cervical cancer;    Random forest model;    Bioinformatics;   

DOI  :  10.1186/s12885-020-07265-7

来源: Springer

【 摘 要 】

BackgroundTo study the risk factors involved in the occurrence and progression of cervical intraepithelial neoplasia (CIN) and to establish predictive models.MethodsGenemania was used to build a gene network. Then, the core gene-related pathways associated with the occurrence and progression of CIN were screened in the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. Real-time fluorescence quantitative polymerase chain reaction (RT-qPCR) experiments were performed to verify the differential expression of the identified genes in different tissues. R language was used for predictive model establishment.ResultsA total of 10 genes were investigated in this study. A total of 30 cases of cervical squamous cell cancer (SCC), 52 cases of CIN and 38 cases of normal cervix were enrolled. Compared to CIN cases, the age of patients in the SCC group was older, the number of parities was greater, and the percentage of patients diagnosed with CINII+ by TCT was higher. The expression of TGFBR2, CSKN1A1, PRKCI and CTBP2 was significantly higher in the SCC groups. Compared to patients with normal cervix tissue, the percentage of patients who were HPV positive and were diagnosed with CINII+ by TCT was significantly higher. FOXO1 expression was significantly higher in CIN tissue, but TGFBR2 and CTBP2 expression was significantly lower in CIN tissue. The significantly different genes and clinical factors were included in the models.ConclusionsCombination of clinical and significant genes to establish the random forest models can provide references to predict the occurrence and progression of CIN.

【 授权许可】

CC BY