Using Graph Databases to Investigate Trends in Structure–Activity Relationship Networks

Abstract

Mining the steadily increasing amount of chemical and biological data is a key challenge in drug discovery. Graph databases offer viable alternatives for capturing interrelationships between molecules and for generating novel insights for design. In a graph database, molecules and their properties are mapped to nodes, while relationships are described by edges. Here, we introduce a graph database for navigation in chemical space, analogue searching, and structure–activity relationship (SAR) analysis. We illustrate this concept using hERG channel inhibitors from ChEMBL to extract SAR knowledge. This graph database is built using different relationships, namely 2D-fingerprint similarity, matched molecular pairs, topomer distances, and structure–activity landscape indices (SALI). Typical applications include retrieving analogues linked by single or multiple edge paths to the query compound as well as detection of nonadditive SAR features. Finally, we identify triplets of linked molecules for clustering. The speed of searching and analysis allows the user to interactively navigate the database and to address complex questions in real-time.

Labels as a Feature: Network Homophily for Systematically Discovering human GPCR Drug-Target Interactions

“…For a similarity threshold of ϵ > 0.4 we compute an all-vs-all similarity metric using Tanimoto similarity, values less than ≤ 0.4 are set to zero to keep a sparse adjacency matrix (otherwise a computer runs into memory limitations). We take inspiration from [12, 21]. Compounds are encoded with a 4096 bit Morgan fingerprint in RDKit (Chirality=True, radius=3) and concatenated with the 166 bit MACCS key substructure fingerprint.…”

“…These network-based methods have found their way to established chemical bioinformatics companies (e.g. Chemaxon with Neo4j 1 ) and have been key tools for SAR campaigns [12]. Preliminary results on successfully utilising CSNs for lead-compound identification were recently demonstrated [15].…”

“…Chemical space networks (CSNs) and similar methods have been developed to rationally engineer drugs by considering the chemical neighbourhood and building structure–activity relationship (SAR) [12]. However, these methods are difficult to scale due to the quadratic dependence on number of compounds and arbitrary design choices in their construction [13, 14].…”

Matched Molecular Pair Analysis in Drug Discovery: Methods and Recent Applications

“…Aside from the traditional optimization rule summary, the MMP concept and method have also been applied in other aspects of drug design, such as scaffold hopping, graph database construction, focused library design, and especially de novo drug design. ,− Most importantly, MMPA can be applied to automatically design compounds with desired drug-like properties and biological activity. In 2019, Green et al reported BRADSHAW, an automated de novo design system that integrates chemical structure generation, experimental design, and active learning.…”

Harnessing Chemical Space Neural Networks to Systematically Annotate GPCR ligands

“…Chemical space networks (CSNs) have been developed to rationally engineer drugs by considering the chemical neighbourhood and building a structure-activity relationship (SAR) (4). However, these methods are difficult to scale due to the quadratic dependence on number of compounds and arbitrary design choices in their construction (5,6).…”