Background: Precise classification of cancer types is critically important for early cancer diagnosis and treatment. Numerousefforts have been made to use gene expression profiles to improve precision of tumor classification. However, reliable cancerrelatedsignals are generally lacking.Method: Using recent datasets on colon and prostate cancer, a data transformation procedure from single gene expression topair-wise gene expression ratio is proposed. Making use of the internal consistency of each expression profiling dataset thistransformation improves the signal to noise ratio of the dataset and uncovers new relevant cancer-related signals (features). Theefficiency in using the transformed dataset to perform normal/tumor classification was investigated using feature partitioningwith informative features (gene annotation) as discriminating axes (single gene expression or pair-wise gene expression ratio).Classification results were compared to the original datasets for up to 10-feature model classifiers.Results: 82 and 262 genes that have high correlation to tissue phenotype were selected from the colon and prostate datasetsrespectively. Remarkably, data transformation of the highly noisy expression data successfully led to lower the coefficient ofvariation (CV) for the within-class samples as well as improved the correlation with tissue phenotypes. The transformed datasetexhibited lower CV when compared to that of single gene expression. In the colon cancer set, the minimum CV decreased from45.3% to 16.5%. In prostate cancer, comparable CV was achieved with and without transformation. This improvement in CV,coupled with the improved correlation between the pair-wise gene expression ratio and tissue phenotypes, yielded higherclassification efficiency, especially with the colon dataset-from 87.1% to 93.5%. Over 90% of the top ten discriminating axes inboth datasets showed significant improvement after data transformation. The high classification efficiency achieved suggestedthat there exist some cancer-related signals in the form of pair-wise gene expression ratio.Conclusion: The results from this study indicated that: 1) in the case when the pair-wise expression ratio transformationachieves lower CV and higher correlation to tissue phenotypes, a better classification of tissue type will follow. 2) thecomparable classification accuracy achieved after data transformation suggested that pair-wise gene expression ratio betweensome pairs of genes can identify reliable markers for cancer.