Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Review
. 2022 Jan 17;23(1):bbab476.
doi: 10.1093/bib/bbab476.

Artificial intelligence in the prediction of protein-ligand interactions: recent advances and future directions

Ashwin Dhakal  1 , Cole McKay  2 , John J Tanner  2   3 , Jianlin Cheng  1
Affiliations
Review

Artificial intelligence in the prediction of protein-ligand interactions: recent advances and future directions

Ashwin Dhakal et al. Brief Bioinform. .

Abstract

New drug production, from target identification to marketing approval, takes over 12 years and can cost around $2.6 billion. Furthermore, the COVID-19 pandemic has unveiled the urgent need for more powerful computational methods for drug discovery. Here, we review the computational approaches to predicting protein-ligand interactions in the context of drug discovery, focusing on methods using artificial intelligence (AI). We begin with a brief introduction to proteins (targets), ligands (e.g. drugs) and their interactions for nonexperts. Next, we review databases that are commonly used in the domain of protein-ligand interactions. Finally, we survey and analyze the machine learning (ML) approaches implemented to predict protein-ligand binding sites, ligand-binding affinity and binding pose (conformation) including both classical ML algorithms and recent deep learning methods. After exploring the correlation between these three aspects of protein-ligand interaction, it has been proposed that they should be studied in unison. We anticipate that our review will aid exploration and development of more accurate ML-based prediction strategies for studying protein-ligand interactions.

Keywords: binding affinity; binding pose; binding site; deep learning; drug discovery; machine learning; protein–ligand interaction.

PubMed Disclaimer

Figures

Figure 1
Figure 1
(A) Sketch of peramivir, an inhibitor of the viral protein neuraminidase from the H1N9 influenza virus. (B) Sketch of human Src kinase inhibitor bosutinib.
Figure 2
Figure 2
Human Src kinase docked by bosutinib visualized with a hydrophobic surface generated in Chimera, PDB code 4MX0. Most hydrophobic regions colored red; most hydrophilic indicated in blue.
Figure 3
Figure 3
Conceptual workflow of ML pipeline. Inputs are the properties of the target protein and ligands, and output are the predicted interactions.
Figure 4
Figure 4
Pie charts showing the distribution of prevailing datasets for the AI-based PLI prediction models. (A) Prevalent dataset for AI based protein-ligand binding affinity prediction models. (B) Prevalent dataset for AI based protein-ligand binding pose prediction models. (C) Prevalent dataset for AI based protein-ligand binding site prediction models.
Figure 5
Figure 5
Statistics of PDBBind dataset showing its composition from Version 2015 as well as the basic structure of Version 2020.
Figure 6
Figure 6
Protein–ligand interactions demonstrated through neuraminidase–peramivir interaction. (A) Neuraminidase monomer with peramivir depicted in red. (B) view of full monomer with peramivir with hydrogen bonding pairs labeled and displayed in canonical atom coloring. Oxygen colored red, and nitrogen in blue. (C) focused view of neuraminidase peramivir hydrogen bonding, PDB code 1L7F.
Figure 7
Figure 7
In silico prediction of two similar analogue Inhibitors docked within the binding sight-2 of human dynamin-1 PH domain. (Figure adopted from [141]) Both analogues share very similar intermolecular forces such as H-bonding and yet slight differences in the ligand’s orientation occur.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Chen K, Mizianty MJ, Kurgan L. ATPsite: sequence-based prediction of ATP- binding residues. IEEE Int Conf Bioinforma Biomed 2010;2011(9):1–8. - PMC - PubMed
    1. Ma B, Shatsky M, Wolfson HJ, et al. . Multiple diverse ligands binding at a single protein site: a matter of pre-existing populations. Protein Sci 2009;11:184–97. - PMC - PubMed
    1. Miller DW, Dill KA. Ligand binding to proteins: the binding landscape model. Protein Sci 1997;6:2166–79. - PMC - PubMed
    1. Whitesides GM, Krishnamurthy VM. Designing ligands to bind proteins. Q Rev Biophys 2005;38:385–95. - PMC - PubMed
    1. Weisel M, Proschak E, Schneider G. PocketPicker: analysis of ligand binding-sites with shape descriptors. Chem Cent J 2007;1:1–17. - PMC - PubMed

Publication types