Screen and classify genes on bladder cancer associated with metastasis

Bladder urothelial carcinoma (BLCA) is a common tumor in humans and a multi-factorial disease. Gene expression may predict the metastasis of cancer. This study aimed to screen differential expression genes related to metastasis of bladder cancer and classify these genes in different groups. The clinical and RNA sequencing (RNA-Seq) expression data in patients with BLCA were downloaded from the TCGA data portal. A total of 407 BLCA samples were available in the TCGA database, for which the clinical information was acquired, of which 407 are RNA-Seq samples with a total of 19 paired samples. SPSS 22.0 was used to find the differentially expressed genes related to bladder cancer with metastasis. Software ‘Gcluto’ was used to cluster the patients with bladder cancer according to the expression of genes and the character of clinical information. Two hundred and fourteen differentially expressed genes related to metastasis of bladder cancer were found from 20,530 genes. Genes were clustered according to the signal pathways, and 19 pairs of patients were clustered according to clinical information and gene expression into 5 groups and the relationships between the genes and the clinical information were found. FBN2 may be related to the papillary subtype of bladder cancer, DEIP1 may be related to the smoking history and the with tumor status of bladder cancer patients. Patients with IER2, C19orf6, CAPZA1, AES, MYL12B and ATPIF1 may have high opportunity of bladder cancer, while CLDND2, PCDHGA6 and SLITRK4 may have less association to bladder cancer compared with other genes.