Análise de bioinformática das neoplasias de glândulas salivares

Abstract

It is important obtain knowledge of the biological processes underlying the developing of tumors of salivary glands with the objective of identify potentials biomarkers or new therapeutic molecular targets. This study was made to investigate the genes leaders of tumors of salivary glands. At first the genes related to all type of tumors of the salivary glands were identified using the databases Pubmed, OMIM and Genecards. Then, the database STRING was used to construct networks of protein-protein interactions to each specific tumor and the values of weighted number of links (WNL) and total interaction score (TIS) were obtained. In this first phase of this study, it was found a total of 3070 genes related with the 35 types of salivary gland tumors. The leaders genes were identified using the algorithm of K-means and Immune/Neural approach. The latter method of data clustering allowed the use of another innovated method which could add or replace more traditional methods like the K-means. Using these two algorithms, 764 genes were identified as leaders genes. And the identification of the genes with the higher values of WNL, in the other words, the genes with more specific biological ways in the tumors in study, and the genes with lower values of TIS, it means that they have a little importance in essential biological process. From the total of leader genes, only 92 leader genes were associated exclusively with benign tumors and 189 others leaders genes were associated exclusively with malignant tumors. Using the last two groups of genes, it was achieved two new networks of protein-protein interactions with the same bioinformatics method, including the calculation of values of WNL and TIS and K-means and Immune/Neural algorithms. The genes UBE2K, CUL1, HGS, UBE2D1, SDHB, UBQLN1, UBE2D2 and RAD23A were identified as leaders genes exclusively of benign tumors. And the genes INS, HRAS, SRC and JUN were established as leader genes exclusively of malignant tumors. Again, from these data, another search for leader gene was proceeded. Then, it was identified the gene RPS15 associated with benign tumors, and the gene PIK3CA in the malignant tumors. These two genes could be possible therapeutic molecular targets or biomarkers. All this data be available in the APPENDIXES A, B and C. With the objective of validation of the hypothesis created using that bioinformatic method, it was achieved experiments with RNA isolation and real time PCR in samples of salivary glands neoplasm that demonstrated higher level of genetic expression of RPS15 in benign tumors in comparison with malignant tumors.