New mathematical measures for apprehending complexity of chiral molecules using information entropy

Patrick Piras (1)

1 ISM2 - Institut des Sciences Moléculaires de Marseille

In this paper, we present several new theoretical measures based on information entropy that can be used to analyze the information content of a chiral molecule. Starting from a differentiation between "chiral" and "achiral" portions in a chiral molecule, we define a new concept that allows us to quantify the complexity of chiral constitutional 2D-isomers of C 10 to C 20 alkanes. Various new chiral and achiral information measures founded on joint entropy, mutual information, and conditional entropy are presented providing an access to a set of regression equations. Then, introducing a case-based measure of entropy, we demonstrate that the distribution of the chiral complexity in these molecules is mostly skewed-right: 60% of the chiral isomers follow a 60/40 distribution rule, which indicates a concentration of chiral complexity in a small number of topological features. Furthermore, by replacing 2D topological distances by 3D distances, the application of these new information measures goes from conformational to racemization and deracemization studies.

Mots clés: chemical graph; chiral alkanes; information theory; molecular diversity; Shannon entropy

Domaines: Chimie; Mathématiques [math]

Patrick Piras. New mathematical measures for apprehending complexity of chiral molecules using information entropy. Chirality, 2022, 34 (4), pp.646-666. ⟨10.1002/chir.23423⟩. ⟨hal-03956655v2⟩