Journal Article

Cycle-configuration descriptors: a novel graph-theoretic approach to enhancing molecular inference

Song, Bowen, Zhu, Jianshen, Akutsu, Tatsuya, Azam, Naveed Ahmed, Haraguchi, Kazuya, Zhao, Liang

2025

Overview

Inference of molecules with desired activities/properties is one of the key and challenging issues in cheminformatics and bioinformatics. For that purpose, our research group has recently developed a state-of-the-art framework mol-infer for molecular inference. This framework first constructs a prediction function for a fixed property using machine learning models, which is then simulated by mixed-integer linear programming to infer desired molecules. The accuracy of the framework heavily relies on the representation power of the descriptors. In this study, we highlight a typical class of non-isomorphic chemical graphs with reasonably different property values that cannot be distinguished by the standard “two-layered (2L) model\" of mol-infer . To address this distinguishability problem of the 2L model, we propose a novel family of descriptors, named cycle-configuration (CC) , which captures the notion of ortho/meta/para patterns that appear in aromatic rings, which was impossible in the framework so far. Extensive computational experiments show that with the new descriptors, we can construct prediction functions with similar or better performance for all 44 tested chemical properties, including 27 regression datasets and 17 classification datasets comparing with our previous studies, confirming the effectiveness of the CC descriptors. For inference, we also provide a system of linear constraints to formulate the CC descriptors as linear constraints. We demonstrate that a chemical graph with up to 50 non-hydrogen vertices can be inferred within a practical time frame. Scientific Contribution This study proposes a new family of descriptors, cycle-configuration ( CC ), for the molecular inference framework mol-infer . Computational experiments demonstrate that incorporating CC descriptors into the 2L model (2L+CC model) can improve the performance of a prediction function in many cases. We also provide an MILP formulation that can infer chemical graphs with up to 50 non-hydrogen atoms within a few minutes.

• Publisher: Springer International Publishing,BioMed Central Ltd,Springer Nature B.V,BMC
• Subject: Analysis / Apexes / Aromatic compounds / Bioinformatics / Chemical properties / Chemistry / Chemistry and Materials Science / Computational Biology/Bioinformatics / Computer applications / Computer Applications in Chemistry / Configurations / Constraints / Datasets / Deep learning / Descriptor design / Documentation and Information in Chemistry / Graph theory / Graphs / Hydrogen / Hydrogen atoms / Inference / Informatics / Integer programming / Inverse QSAR/QSPR / Linear programming / Machine learning / Mixed integer / Mixed integer linear programming / Molecular inference / Performance enhancement / Predictions / Property values / Real estate / Theoretical and Computational Chemistry