Quantum machine learning framework for virtual screening in drug discovery: a prospective quantum advantage

Abstract Machine Learning for ligand based virtual screening (LB-VS) is an important in-silico tool discovering new drugs in a faster and cost-effective manner, especially emerging diseases such as COVID-19. In this paper, we propose general-purpose framework combining classical Support Vector Classifier algorithm with quantum kernel estimation LB-VS on real-world databases, argue favor of its prospective advantage. Indeed, heuristically prove that our integrated workflow can, at least some relevant instances, provide tangible advantage compared to state-of-art algorithms operating the same datasets, showing strong dependence target features selection method. Finally, test IBM Quantum processors using ADRB2 COVID-19 hardware simulations results line predicted performances can surpass equivalents.