Personalized Metabolic Analysis of Diseases

The metabolic wiring of cells is altered drastically in many diseases. Understanding the nature of such changes may pave the way for new therapeutic opportunities, and the development of personalized treatment strategies. In this research, we developed an algorithm, Metabolitics (Cakmak and Celik, 2020), which allows systems-level analysis of changes in the biochemical network of cells in disease states. We demonstrate the use of Metabolitics on three distinct diseases, namely, breast cancer, Crohn’s disease, and colorectal cancer. Our results show that the constructed machine learning models successfully diagnose patients by over 90% accuracy on average. Moreover, we filed an international patent (Cakmak and Celik, 2018) that describes the use of Metabolitics as part of a medical decision support system to help clinicians diagnose diseases in a low-cost and non-invasive manner. We also developed a web-based tool, MetaboliticsDB (Celik et al., 2020), that makes Metabolitics available to researchers from all around the world. It allows users to analyze their Metabolomics data and apply the pre-built machine learning models to identify potential associated diseases with advanced visualization support.