MDACC-RPPA-Analysis

Created By Zhenlin Ju zju
PanCan11_RBN_RPPA_without_Duplicates_20130325.csv Contains protein expression data for 131 proteins and 3467 pancan11 tumor samples, measured by RPPA (reverse phase protein array) technology. These data have been normalized by RBN (replicate-base normalization) method developed by Dr. Rehan Akbani. PanCan11_RBN_SimpleCluster_20130411.csv Contains TCGA ID and cluster assignments (K=8) of Pan-cancer 11 samples. ZJU_AdjCl_PanCan11_TCGA_RBN_RPPA_Heatmap_Simple_K8_V4_1365710010900.pdf Heat map showing the Pan-cancer 11 samples clusters (K=8) and other annotation bars. RBN k = 8 Tumor Colors Updated 7/11/2013 ZJu: pancan11-131P-RBN-NGCHM PDF: syn1971263 pancan11_driver_proteins.xlsx Driver proteins for each cluster were obtained by using the Information Gain criterion. Information Gain was computed for each protein for each cluster by dichotomizing the data into the cluster of interest vs. all other clusters. A threshold of 0.05 was then applied and only proteins with Information Gain >= 0.05 were retained in the list of drivers. Users may decide to use higher thresholds if they want. The Information Gain value for each protein in each cluster is given in the file. Direction was computed by comparing the mean value of the protein in the cluster of interest vs. the mean value in all other clusters. A direction of ?high? means the mean value was higher in the cluster of interest, and a direction of ?low? means vice versa.

syn1870500
syn1759392
syn1756922
syn1756921
syn1971263
syn1773107
syn1870498