Utilizing graph machine learning within drug discovery and development

Abstract

Graph machine learning (GML) is receiving growing interest within the pharmaceutical and biotechnology industries for its ability to model biomolecular structures, the functional relationships between them, and integrate multi-omic datasets — amongst other data types. Herein, we present a multidisciplinary academic-industrial review of the topic within the context of drug discovery and development. After introducing key terms and modelling approaches, we move chronologically through the drug development pipeline to identify and summarize work incorporating: target identification, design of small molecules and biologics, and drug repurposing. Whilst the field is still emerging, key milestones including repurposed drugs entering in vivo studies, suggest GML will become a modelling framework of choice within biomedical machine learning.

A human-based multi-gene signature enables quantitative drug repurposing for metabolic disease

“…Publicly available chemogenomic databases are very far from complete, and therefore, ML modelling approaches can be used to provide estimates for missing data. DL-based models have shown promise in this context ( Gaudelet et al, 2021 ; James et al, 2020 ). We used DeepPurpose, a DL library for DTI prediction ( Huang et al, 2021 ) that takes as an input SMILES of the small molecules of interest and the amino acid sequences of the protein-coding genome.…”

Applied machine learning in Alzheimer's disease research: omics, imaging, and clinical data

“…By identifying hidden drug–target interaction, it is possible to identify existing drugs that may have new indications for AD. Hidden drug–target interaction can be revealed by finding new drug–target binding (off-target) or by integrating multi-modal interactions (on-target) [ 103 ]. The off-target approach uses biochemical properties (structural, ligand-based molecular docking) or biophysical properties (3D conformation) to predict drug–target binding [ 85 , 104 ].…”

Graphein - a Python Library for Geometric Deep Learning and Network Analysis on Protein Structures and Interaction Networks

“…Biomolecular entities can be represented as nodes, and their associated functional relationships and physical interactions can be represented as edges with associated metadata, such as the direction and nature of regulation. For a full discussion of applications, datasets and modelling techniques we refer readers to the reviews by [44] and [9]. Graphein implements interaction graph construction from protein-protein interaction and gene regulatory network databases.…”

“…Geometric deep learning refers to the application of deep learning methods to data with an underlying non-Euclidean structure, such as graphs or manifolds [1]. These methods have already been applied to a number of problems within computational biology and computational structural biology [2, 3, 4, 5, 6, 7, 8], and have shown great promise in the contexts of drug discovery and development [9]. Geometric deep learning libraries have been developed, providing graph representation functionality and in-built datasets - typically with a focus on small molecules [10, 11].…”