Extracting dynamics from static cancer expression data

This webpage provides supporting information for the IEEE/ACM Transactions on Computational Biology and Bioinformatics 2007 submission by A. Gupta and Z. Bar-Joseph.


Polynomial analysis

Confidence analysis

Survival for glioma ordering

GO p-values for genes

Oncogenes and tumor suppressors

Matlab implementation

This website provides additional information and figures that were omitted from our submission due to lack of space. Follow the link on the left to view the results for the polynomial synthetic data, error bars for comparisons between ordering methods, the orderings determined for the two glioma datasets, the GO enrichment tables for the genes displaying consistent behavior in the ordering and plots for oncogenes and tumor suppressors. A Matlab implementation can also be downloaded.

Cancer survival times (in days) based on the ordering recovered by our algorithm. Note that some of the patients were alive when the data was analyzed. More living patients are assigned to the right half and patients on that side lived longer. One possible explanation is that they were diagnosed earlier in their disease indicating that the ordering produced by our algorithm is reasonable. This plot presents the ordering recovered for the entire test set.