A method and a system are provided for determining a likelihood of a metabolic pathway existing in an organism, including: calculating a feature vector for each metabolic pathway of metabolic pathways known to exist and not known to exist in the organism, wherein elements of the feature vectors are network properties of the metabolic pathways mapped to a metabolite concentration correlation network (CN); training a supervised machine learning model for classifying metabolic pathways as existing or not existing in the organism, according to the known and unknown feature vectors; determining, based on mapping to the CN, a feature vector of a proposed metabolic pathway, and feeding the feature vector of the proposed metabolic pathway to the trained SML model, to determine a likelihood of the proposed metabolic pathway existing in the organism.