Can “Easy” Chemistry Produce Complex, Diverse and Novel Molecules?

Abstract

<p>Prompted by reoccurring statements found in the literature, at seminars, and at conferences, regarding the perceived inferiority of “easy-to-make” compounds compared to molecules obtained through other reaction schemes, we have investigated if there are significant differences between these two groups of compounds. As a surrogate for the “easy-to-make” set, we compiled screening compounds made from the most popular chemical reaction in medicinal chemistry, amide bond formation. The other set contained screening compounds made by any other coupling reaction using the expert system for named reaction identification and classification (namerxn) from NextMove Scientific Software. The data sets were rigorously compiled from the AstraZeneca electronic notebooks and screening databases. We can conclude that novelty, diversity and molecular complexity is nowadays not compromised by the use of amide bond formation, mainly due to a steady increase of unique amines made available during the last decades. With this, we can demonstrate that “easy” chemistry allows speedy access to a broad chemical space, facilitating progress in medicinal chemistry projects, and opens the possibility of synthesis automation and new technologies like DNA encoded libraries. That said, we also stress that innovative chemistry must be continuously developed to maintain the production of new elaborate building blocks enabling fast advancement in medicinal chemistry and drug design. </p>

Generating Multibillion Chemical Space of Readily Accessible Screening Compounds

“…One might argue that the technology described in the current work is mostly based on very simple chemical transformations; therefore, its capability of producing novel, complex, and diverse molecules might be questionable. Nevertheless, a recent analysis by AstraZeneca scientists shows that this is not the case: even using only the amide formation reaction, very good results can be obtained in the early drug discovery provided that sufficient access to the corresponding building blocks is possible ( Tomberg and Boström, 2020 ).…”

MEMES: Machine learning framework for Enhanced MolEcular Screening

“…On the contrary to the molecules generated by deep generative models, molecule library enumerated via simple reactions can be novel, diverse and at the same time practically be synthesized with a probability of ∼ 86%. [30][31][32] In a recent study performed by Lyu et al, 31 96 million docking calculations were performed against AmpC receptor. Among these the top ranked 1 million compounds ( 1 % of the initial set) were systematically examined to identify hit molecules, which were further validated experimentally.…”