ChemLog: Making MSOL Viable for Ontological Classification and Learning

TL;DR

This paper introduces ChemLog, a method combining monadic second-order logic formalizations with deep learning to improve ontology classification in chemistry, enabling scalable and more accurate peptide class identification.

Contribution

It presents a novel approach integrating logical formalizations with transformer models to enhance ontology classification in chemistry.

Findings

Logical formalization applied to peptide classes in ChEBI and PubChem.

Deep learning models' performance improved using logic-based classifications as training data.

Scalable classification across large chemical databases achieved.

Abstract

Despite its prevalence, in many domains, OWL is not expressive enough to define ontology classes. In this paper, we present an approach that allows to use monadic second-order formalisations for ontology classification. As a case study, we have applied our approach to 14 peptide-related classes from the chemistry ontology ChEBI. For these classes, a monadic second-order logic formalisation has been developed and applied both to ChEBI as well as to 119 million molecules from the chemistry database PubChem. While this logical approach alone is limited to classification for the specified classes (in our case, (sub)classes of peptides), transformer deep learning models scale classification to the whole of the ChEBI ontology. We show that when using the classifications obtained by the logical approach as training data, the performance of the deep learning models can be significantly enhanced.