Accelerating Drug Discovery with Quantum Computing: Opportunities and Challenges

Abstract

The increasing complexities in discovering drugs require computational framework that would address problems that are beyond the capabilities of classical computing. The conventional drug development pipeline is time and resource-intensive, where the anticipated length is more than ten years and approximately 2.6 billion dollars is spent developing a profitable drug (DiMasi et al., 2016). Quantum computation has become a promising paradigm, which has the prospect to create a representation of the quantum of scientifically complex molecular systems with exponentially high precision and effectiveness (Cao et al., 2019). How quantum computing can be combined and used with artificial intelligence (AI) and machine learning (ML) models and speed up different stages of the drug discovery pipeline, including molecular docking to optimization, is investigated in this study (Biamonte et al., 2017). To study this, using SBM, we find that it may be useful in the case of solving NP-hard problems involved in molecular modelling by utilising the hybrid quantum classical algorithm, such as Variational Quantum Eigensolvers (VQE) and Quantum Approximate Optimization Algorithms (QAOA) (Peruzzo et al., 2014). There are always opportunities to improve prediction performance and optimization and a lower computational load, but there are also critical tasks, including noise, error correction, and scalability to hardware (Preskill, 2018). The paper ends with a road map to practical application and prospects of future research where the importance of interdisciplinary collaboration is stated in order to address current shortcomings and implement quantum advantage in real pharmaceutical results (McArdle et al., 2020).

Keywords: quantum compute machine learning, drug discivery, molecular simulation, AI, quantum algorit, quantum computing