Efficient and effective exploration of molecular shapes for drug discovery

Taking a step back in time highlights that drug discovery can be a peculiar and difficult process to control. A well-known example of this includes Alexander Fleming's discovery of penicillin in 1928, when one of his staphylococcus culture plates became contaminated and developed a mould that created a bacteria-free circle. Another famous example stems back to 1943 when the chemist Albert Hoffman discovered the psychotomimetic effect of lysergic acid diethylamide (LSD-25) via accidental consumption[1]. Although brief, these examples reveal that serendipity has played a significant role in the realm of drug discovery throughout history. One paper investigating the role of serendipity in anticancer drug discovery, states that 5.8% of all the drugs within the market were discovered as a result of serendipity[2]. However, it is not just good fortune that the pharmaceutical industry must overcome when tasked with discovering potential drug compounds. A study has shown that the task of discovering a new drug can take between 10 - 17 years[3] and the findings of another study has highlighted that the average cost of developing a new drug is 1.3 billion dollars[4]. Furthermore, this process involves the synthesising and testing of thousands of possible candidates. However, chemical synthesis provides access to a monolithic number of drug-like molecules, which can be estimated to be in the order of thousands of billions and understandably far exceeds the capacity of experimental screening techniques, such as high-throughput screening.