Support vector machine classifier for prediction of the metastasis of colorectal cancer

Colorectal cancer (CRC) is one of the most common cancers and a major cause of mortality. The present study aimed to identify potential biomarkers for CRC metastasis and uncover the mechanisms underlying the etiology of the disease. The five datasets GSE68468, GSE62321, GSE22834, GSE14297 and GSE6988 were utilized in the study, all of which contained metastatic and non-metastatic CRC samples. Among them, three datasets were integrated via meta-analysis to identify the differentially expressed genes (DEGs) between the two types of samples. A protein-protein interaction (PPI) network was constructed for these DEGs. Candidate genes were then selected by the support vector machine (SVM) classifier based on the betweenness centrality (BC) algorithm. A CRC dataset from The Cancer Genome Atlas database was used to evaluate the accuracy of the SVM classifier. Pathway enrichment analysis was carried out for the SVM-classified gene signatures. In total, 358 DEGs were identified by meta-analysis. The top ten nodes in the PPI network with the highest BC values were selected, including cAMP responsive element binding protein 1 (CREB1), cullin 7 (CUL7) and signal sequence receptor 3 (SSR3). The optimal SVM classification model was established, which was able to precisely distinguish between the metastatic and non-metastatic samples. Based on this SVM classifier, 40 signature genes were identified, which were mainly enriched in protein processing in endoplasmic reticulum (e.g., SSR3), AMPK signaling pathway (e.g., CREB1) and ubiquitin mediated proteolysis (e.g., FBXO2, CUL7 and UBE2D3) pathways. In conclusion, the SVM-classified genes, including CREB1, CUL7 and SSR3, precisely distinguished the metastatic CRC samples from the non-metastatic ones. These genes have the potential to be used as biomarkers for the prognosis of metastatic CRC.