The use of computational biology in protein engineering and drug discovery

The traditional method of drug discovery can take up to 12 years to make a novel drug. Conventional means of drug discovery are costly and time-consuming; this leads to high drug prices, making novel treatments less affordable for patients. Using machine learning and deep learning algorithms can help increase research efficiency, decrease associated costs, and improve the affordability of future marketed drugs. Computer-aided drug discovery includes a large set of tools widely used to develop and optimize drug-like small molecules based on the target's molecular structure analysis. For instance, these methods can analyze the similarities between the molecular structure of the target and the available data to develop a small molecule with relevant structural properties. The drug discovery research and development process can be largely automated using machine learning and artificial intelligence techniques. However, the limited access to curated and properly annotated data has partially impaired the deployment of these data-driven techniques. In this chapter, we have screened thousands of recent publications in the drug discovery and development field to identify recent advances in computer-aided drug discovery. We review the recent progress within a selected set of techniques and discuss how these methods position themselves as game changers in early-stage drug discovery.