Breast cancer data from TCGA for machine learning exercise

I accessed the data used in this publication (tcga-data.nci.nih.gov/docs/publications/brca_2012/) and parsed the clinical data to determine which patients were HER2 positive. I then extracted gene-expression data for these patients and created a matrix that indicates expression levels for all genes with no missing values. The first row indicates HER2 status (1 = positive, -1 = negative)