BACKGROUND: The overall survival rate of ovarian cancer patients is still poor because of the difficulties encountered in detection, diagnosis and treatment. Here, we aim to systematically identify the genetic factors causing ovarian cancer and find the accurate diagnostic and therapeutic targets for ovarian cancer. METHODS: We collected the known archived ovarian cancer-related genes from the databases used as the investigated targets and employed the minimum redundancy maximum relevance and random forest classification to identify the novel ovarian cancer-related genes in addition to the known ones. We further identified candidates as the markers for the detection of the ovarian cancer based on the gene expression data and then confirmed them by quantitative real-time PCR. RESULTS: We found out the genetic terms to interpret the mechanism of ovarian cancer. Based on those terms, we predicted 860 novel related genes as candidates. These candidates can act as expression biomarkers for clinical detection and they achieved a 100% accuracy. We verified 10 of them as the optimal biomarkers for detection in the expression data. CONCLUSION: We employed the features of achieved ovarian cancer-related genes to identify 860 novel ovarian cancer genes. We further validated 10 genes as biomarkers for detection of ovarian cancer.
BACKGROUND: The overall survival rate of ovarian cancerpatients is still poor because of the difficulties encountered in detection, diagnosis and treatment. Here, we aim to systematically identify the genetic factors causing ovarian cancer and find the accurate diagnostic and therapeutic targets for ovarian cancer. METHODS: We collected the known archived ovarian cancer-related genes from the databases used as the investigated targets and employed the minimum redundancy maximum relevance and random forest classification to identify the novel ovarian cancer-related genes in addition to the known ones. We further identified candidates as the markers for the detection of the ovarian cancer based on the gene expression data and then confirmed them by quantitative real-time PCR. RESULTS: We found out the genetic terms to interpret the mechanism of ovarian cancer. Based on those terms, we predicted 860 novel related genes as candidates. These candidates can act as expression biomarkers for clinical detection and they achieved a 100% accuracy. We verified 10 of them as the optimal biomarkers for detection in the expression data. CONCLUSION: We employed the features of achieved ovarian cancer-related genes to identify 860 novel ovarian cancer genes. We further validated 10 genes as biomarkers for detection of ovarian cancer.