Skip to main content
eScholarship
Open Access Publications from the University of California

Support Vector Machine for Kidney Cancer Classification

  • Author(s): Cao, Liyu
  • Advisor(s): Brody, James
  • et al.
Abstract

Renal cancer is the 12th leading cause of cancer death, accounting for 2.4 percent of all cancers in the United States. Two of most common types of kidney tumors are clear cell and papillary carcinoma. In this study, data of Kidney Renal Clear Cell Carcinoma (KIRC) and Kidney renal papillary cell carcinoma (KIRP) projects from the Cancer Genome Altas database were analyzed. These two data were combined and most part of it was used as a training set for candidate biomarker identification and 50 of the dataset was used to test the accuracy of classification model based on the identified biomarker. As a result, the accuracy of classification was 96.0% when the model based on top 5 biomarkers was trained by gene expression data. For DNA methylation level, the performance of model based on top biomarkers was the test with an overall accuracy of 100.0%. In conclusion, some informative biomarkers that could accurately discriminate KIRC and KIRP were identified in this study.

Main Content
Current View