Renal cell carcinoma classification: a case study of pitfalls associated with metabolic landscape analysis

The huge amount of biological data gained from many different sources opens up new possibilities alongside with some dangers in data integration. The latter are not so widely taken account for. In the present article, we show what problems may arise with integration of metabolomic data with gene expression datasets. We analyze common pitfalls and provide solutions, exemplifying them by a case study of the renal cell carcinoma (RCC). In particular we provide a metabolic description of known morphological RCC subtypes and suggest a possible existence of poor-prognosis cluster of patients, which are characterized by common low activity of drug transporting enzymes important in chemotherapy. Finally, the goal of this work is also to point out the problem that arises from integration of high-throughput data with inherently biased manually curated low-throughput data. In such cases overrepresented information may potentially overshadow the non-trivial discoveries.