To treat or cure a disease, drug discovery is the first link for any pharmaceutical research and development. Identification, optimisation, synthesis and validation using various cell assays and animal models is the first stage of a long journey that a drug candidate undergoes. This initial discovery generally can take anywhere from five to seven years.
Drug discovery is a long and complex process that can be broadly divided into four major stages:(i) target selection and validation; (ii) compound screening and lead optimization; (iii) preclinical studies; and (iv) clinical trials.
First, the target related to a specific disease needs to be identified which requires cellular and genetic target evaluation, genomic and proteomic analysis, and bioinformatic predictions. The next step is hit identification, where compounds are identified from molecular libraries by using methods such as combinatorial chemistry, high-throughput screening, and virtual Structure-activity. In silico studies in combination with cellular functional tests are used in an iterative cycle to improve the functional properties of newly synthesized drug candidates. Subsequently, in-vivo studies such as pharmacokinetic investigations and toxicity tests are performed in animal models. Finally, the drug candidate, which has now successfully passed all preclinical tests, is administered to patients in a clinical trial. It has been estimated that the average cost of a traditional drug discovery pipeline is 2.6 billion USD, and a complete traditional workflow can take over 12 years
How to decrease the costs and speed up projects are central questions for all pharmaceutical companies. AI-based methods are increasingly being used in various stages of the process to improve time and cost-efficiency. These include the use of AI in real-time image-based cell sorting, cell classification, quantum mechanics calculation of compound properties, computer-aided organic synthesis, designing new molecules, developing assays, predicting the 3D structures of target proteins, and many others. In general, these processes are somewhat tedious to perform and can, with the help of AI, be automated and optimized to substantially speed up the R&D drug discovery process. We review below the different subareas of the drug discovery process which have benefitted from incorporating AI.