Personalized discovery of altered pathways in clear cell renal cell carcinoma using accumulated normal sample data

To identify altered pathways in an individual with clear cell renal cell carcinoma (ccRCC) using accumulated normal sample data.

Gene expression data of E-GEOD-40435 was downloaded from the ArrayExpress database. Gene-level statistics of genes in tumor and normal samples were computed. Then, the Average Z method was applied to calculate the individual pathway aberrance score (iPAS). Subsequently, the significantly altered pathways in a ccRCC sample were identified using T-test based on the pathway statistics values of normal and ccRCC samples. Moreover, the identified altered pathways were verified through two methods: one was assessing classification capability for microarray data samples, and the other was computing the changed percentage of each pathway in ccRCC samples.

Based on the threshold, 886 altered pathways were identified in all samples. The most significant pathways were potassium transport channels, proton-coupled monocarboxylate transport, beta oxidation of octanoyl-CoA to hexanoyl-CoA, antigen presentation: folding, assembly and peptide loading of class I MHC, and so on. Additionally, iPAS separated ccRCC from normal controls with an accuracy of 0.980. Moreover, a total of 5 significant pathways with change in 100% ccRCC samples were extracted including proton-coupled monocarboxylate transport, antigen presentation: folding, assembly and peptide loading of class I MHC, and so on.

iPAS is useful to predict marker pathways for ccRCC with a high accuracy. Pathways of proton-coupled monocarboxylate transport, and antigen presentation: folding, assembly and peptide loading of class I MHC might play crucial roles in ccRCC progression.