A Network Inference Method for Large-Scale Unsupervised Identification of Novel Drug-Drug Interactions

Characterizing interactions between drugs is important to avoid potentially harmful combinations, to reduce off-target effects of treatments and to fight antibiotic resistant pathogens, among others. Here we present a network inference algorithm to predict uncharacterized drug-drug interactions. Our algorithm takes, as its only input, sets of previously reported interactions, and does not require any pharmacological or biochemical information about the drugs, their targets or their mechanisms of action. Because the models we use are abstract, our approach can deal with adverse interactions, synergistic/antagonistic/suppressing interactions, or any other type of drug interaction. We show that our method is able to accurately predict interactions, both in exhaustive pairwise interaction data between small sets of drugs, and in large-scale databases. We also demonstrate that our algorithm can be used efficiently to discover interactions of new drugs as part of the drug discovery process.Author Summary: Over one in four adults older than 57 in the US take five or more prescriptions at the same time; as many as 4% are at risk of a major adverse drug-drug interaction. Potentially beneficial effects of drug combinations, on the other hand, are also important. For example, combinations of drugs with synergistic effects increase the efficacy of treatments and reduce side effects; and suppressing interactions between drugs, in which one drug inhibits the action of the other, have been found to be effective in the fight against antibiotic-resistant pathogens. With thousands of drugs in the market, and hundreds or thousands being tested and developed, it is clear that we cannot rely only on experimental assays, or even mechanistic pharmacological models, to uncover new interactions. Here we present an algorithm that is able to predict such interactions. Our algorithm is parameter-free, unsupervised, and takes, as its only input, sets of previously reported interactions. We show that our method is able to accurately predict interactions, even in large-scale databases containing thousands of drugs, and that it can be used efficiently to discover interactions of new drugs as part of the drug discovery process.