This presentation brings together the concept of molecular complexity and crowdsourcing. An exercise was done at Merck where 386 chemists voted on the molecular complexity (on a scale of 1-5) of 2,681 molecules taken from various sources: public, licensed, and in-house. The Mean_Complexity of a molecule is the average over all votes for that molecule. As long as enough votes are cast per molecule, we find Mean_Complexity is quite easy to model with QSAR methods using only a handful of physical descriptors (e.g. number of chiral centers, number of unique topological torsions, Weiner index). The high level of self-consistency (cross-validated R2 ~0.88) is remarkable given that our chemists do not agree with each other strongly about the complexity of any given molecule. Thus, the power of crowdsourcing is clearly demonstrated in this case. Mean_Complexity appears to be correlated with at least one metric of synthetic complexity from the literature derived in a different way and is also correlated with values of process mass intensity (PMI) from the literature and from in-house studies. It is demonstrated that this Complexity can be used to differentiate between medicinal chemistry projects and to follow the chemistry trajectory in projects over time.

Modeling a crowdsourced definition of molecular complexity

248th National Meeting (2014)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.