Model-based identification of drug targets that revert disrupted metabolism and its application to ageing.

The growing availability of 'omics' data and high-quality in silico genome-scale metabolic models (GSMMs) provide a golden opportunity for the systematic identification of new metabolic drug targets. Extant GSMM-based methods aim at identifying drug targets that would kill the target cell, focusing on antibiotics or cancer treatments. However, normal human metabolism is altered in many diseases and the therapeutic goal is fundamentally different--to retrieve the healthy state. Here we present a generic metabolic transformation algorithm (MTA) addressing this issue. First, the prediction accuracy of MTA is comprehensively validated using data sets of known perturbations. Second, two predicted yeast lifespan-extending genes, GRE3 and ADH2, are experimentally validated, together with their associated hormetic effect. Third, we show that MTA predicts new drug targets for human ageing that are enriched with orthologs of known lifespan-extending genes and with genes downregulated following caloric restriction mimetic treatments. MTA offers a promising new approach for the identification of drug targets in metabolically related disorders.