A Review on the Significance of Artificial Intelligence in Drug Discovery and Pharmacology

Abstract

Drug discovery is associated with enormous expenses in terms of financial resources required to introduce one innovative medicine into market. For instance, introducing new drugs is estimated to cost more than $2.6 billion dollars while the process may take up to 10-15 years and 90% of drugs fail during clinical trials. There is another challenge associated with drug discovery. Bacteria are developing increasing resistance to current antibiotics. Therefore, there is an urgent need to develop innovative approaches to drug discovery and development, which have been facilitated by the rapid development of artificial intelligence. Machine learning, deep learning, natural language processing and reinforcement learning are some applications of AI relevant for pharmaceutical industry. Those methods can be used for target identification and validation, molecular designing, virtual screening and docking, ADMET prediction, clinical trials, etc. This systematic review aims to analyze the AI applications in the process of drug discovery. It becomes evident from this systematic review that artificial intelligence has become an integral part of drug discovery and pharmacology. The existing literature have demonstrated some improvements in drug discovery process, achieved by using AI methods such as discovery speed, hit rates, ADMET prediction accuracy, or clinical trial optimization, among others. While there are various challenges to overcome in the field of AI-assisted drug discovery, such as data biases, lack of interpretability, experimental validation gap, or regulatory considerations, the best way going forward is to look at AI as an augmentative intelligence, rather than replacement for medicinal chemists.