Discovery of Boolean metabolic networks: integer linear programming based approach

Abstract

BackgroundTraditional drug discovery methods focused on the efficacy of drugs rather than their toxicity. However, toxicity and/or lack of efficacy are produced when unintended targets are affected in metabolic networks. Thus, identification of biological targets which can be manipulated to produce the desired effect with minimum side-effects has become an important and challenging topic. Efficient computational methods are required to identify the drug targets while incurring minimal side-effects.ResultsIn this paper, we propose a graph-based computational damage model that summarizes the impact of enzymes on compounds in metabolic networks. An efficient method based on Integer Linear Programming formalism is then developed to identify the optimal enzyme-combination so as to minimize the side-effects. The identified target enzymes for known successful drugs are then verified by comparing the results with those in the existing literature.ConclusionsSide-effects reduction plays a crucial role in the study of drug development. A graph-based computational damage model is proposed and the theoretical analysis states the captured problem is NP-completeness. The proposed approaches can therefore contribute to the discovery of drug targets. Our developed software is available at “http://hkumath.hku.hk/~wkc/APBC2018-metabolic-network.zip”.

George Bernard Dantzig

“…Such optimization procedures have not just reduced costs and increased outputs across the economy, they have also saved countless lives. Some recent examples using linear programming include phylogenetic analysis to determine the origins of organisms, such as viruses (Han et al., 2019), electrical stimulation therapy (Abouelseoud et al., 2018), chemotherapy network design (Battesini et al., 2018), discovery of drug targets (Qiu et al., 2018), designs for radiation therapy (Sjölund et al., 2019), and finding optimal diets (Yu et al., 2018), which continues to be an intense area of application starting with Dantzig's original solution (Dantzig, 1990) of the diet problem, a model proposed by the economist George Stigler (1945).…”