VISUAL AND STATISTICAL-BASED CROSS-PLATFORM NORMALIZATION ON GENE EXPRESSION DATA OF ORAL CANCER

Abstract

Oral squamous cell carcinoma is the sixth most common cancer worldwide. The increasing epidemiological relevance of this cancer emphasizes the need to identify predictive tumor markers. There are limited studies to associate the expression changes in OSCC using clinically relevant variables. Many studies showed inconsistent cancer biomarkers due to bioinformatics artifacts. In this work we use multiple data sets from microarrays in order to improve the reliability of cancer biomarkers. Combining a large number of gene expression datasets originating from different labs could be beneficial for the discovery of new biological insights and could increase the statistical power of gene expression analysis, but then this data should be combined in a consistent manner. We perform a Cross-Platform Normalization method which integrates and cross-annotates multiple data sets related to oral cancer. This Cross-Platform Normalization was done to determine differential gene expression in oral cancer using the open-source R programming environment in conjunction with the open-source Bioconductor software. Cross-Platform Normalization is a powerful tool for analyzing microarray experiments by combining data from multiple studies. Functionalities for combining outputs from different methods and for data transformation are also available in the package. Moderate t-statistics is used to find DEG using Limma package of Bioconductor. In this microarray analysis expression profile of samples were used to identify DEG using 352 samples of which 69 was normal while 283 was tumor. Total 16 genes are found to be differentially expressed, seven genes are found to be upregulated (MMP1, MMP12, CXCL8, SPP1, PTHLH, MMP3 and MMP10), while nine genes are found to be downregulated (ENDOU, MAL, CRNN, SCEL, TGM3, CLCA4, KRT4, CRISP3 and KRT13). All these genes are previously shown to be involved in OSCC and hence, can be used as the potential biomarker to detect oral cancer.