A Turing Test for Molecular Generators

Abstract

Machine learning approaches promise to accelerate and improve success rates in medicinal chemistry programs by more effectively leveraging available data to guide a molecular design. A key step of an automated computational design algorithm is molecule generation, where the machine is required to design high-quality, drug-like molecules within the appropriate chemical space. Many algorithms have been proposed for molecular generation; however, a challenge is how to assess the validity of the resulting molecules. Here, we report three Turing-inspired tests designed to evaluate the performance of molecular generators. Profound differences were observed between the performance of molecule generators in these tests, highlighting the importance of selection of the appropriate design algorithms for specific circumstances. One molecule generator, based on match molecular pairs, performed excellently against all tests and thus provides a valuable component for machine-driven medicinal chemistry design workflows.

Schistosomiasis Drug Discovery in the Era of Automation and Artificial Intelligence

“…Despite the innovation of generative models, the novelty and accessibility of generated molecules must be evaluated (372)(373)(374). Gao and Coley (375) observed that generative models can produce infeasible molecules even with good performance in benchmarks.…”

Individual and collective human intelligence in drug design: evaluating the search strategy

“…Until today only timid attempts have been made to address drug design using crowdsourcing. Recently some trials were done by integrating many experts in order to: enhance chemical libraries through the “wisdom of crowds” [ 44 ], model molecular complexity from a crowdsourced medicinal chemist perspective [ 45 ], predict solubility in place of machines [ 46 ], and assess quality of molecules generated by automatic algorithms in Turing-inspired tests [ 47 ]. All such activities are related to scoring strategies of de novo drug design but no endeavor has been made (as far as we know) to deal with the other two elements: the assembly and the search strategies.…”

Individual & Collective Human Intelligence in Drug Design: Evaluating the Search Strategy

“…https://www.ncbi.nlm.nih.gov/pccompound?term=all %5Bfilt%5D&cmd=search b As 27/11/2020 CAS registry contains more than 171 million unique organic and inorganic chemical substances (https://www.cas.org/support/documentation/chemical-substances) some trials were done by integrating many experts in order to: enhance chemical libraries through the "wisdom of crowds", 44 model molecular complexity from a crowdsourced medicinal chemist perspective 45 , predict solubility in place of machines, 46 and assess quality of molecules generated by automatic algorithms in Turing-inspired tests. 47 All such activities are related to scoring strategies of de novo drug design but no endeavor has been made (as far as we know) to deal with the other two elements: the assembly and the search strategies.…”