Hybridization of Chemoinformatic and Bioinformatic Information Networks for Graph Diffusion of Drug-Target Interactions

BRUNO KAUFMAN; A. CHERNOMORETZ

Congreso; NetSciX 2023; 2023

Network Science Society

 Drug development is a costly, time-consuming process. It is therefore convenient to repurpose existing drugsfor novel treatments. Massive amounts of information exist on the topic of drug-protein interactions. In orderto extract new insights from this wealth of data, massive computational approaches are warranted, combiningbioinformatics and chemoinformatics. In our work, we explore the combination of several sources of data: drug-protein interactions and chemical similarity measures based on molecular fingerprints (Tanimoto similarity) anddrug scaffold hierarchies. These come in the form of complex networks, and a hybrid network is obtainedby combining them. In this contribution we present a network diffusion strategy to predict new drug-proteininteractions.