skip to main content


Title: Artificial intelligence in the prediction of protein–ligand interactions: recent advances and future directions
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.  more » « less
Award ID(s):
1759934
NSF-PAR ID:
10332265
Author(s) / Creator(s):
; ; ;
Date Published:
Journal Name:
Briefings in Bioinformatics
Volume:
23
Issue:
1
ISSN:
1467-5463
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Recently, machine learning (ML) has established itself in various worldwide benchmarking competitions in computational biology, including Critical Assessment of Structure Prediction (CASP) and Drug Design Data Resource (D3R) Grand Challenges. However, the intricate structural complexity and high ML dimensionality of biomolecular datasets obstruct the efficient application of ML algorithms in the field. In addition to data and algorithm, an efficient ML machinery for biomolecular predictions must include structural representation as an indispensable component. Mathematical representations that simplify the biomolecular structural complexity and reduce ML dimensionality have emerged as a prime winner in D3R Grand Challenges. This review is devoted to the recent advances in developing low-dimensional and scalable mathematical representations of biomolecules in our laboratory. We discuss three classes of mathematical approaches, including algebraic topology, differential geometry, and graph theory. We elucidate how the physical and biological challenges have guided the evolution and development of these mathematical apparatuses for massive and diverse biomolecular data. We focus the performance analysis on protein–ligand binding predictions in this review although these methods have had tremendous success in many other applications, such as protein classification, virtual screening, and the predictions of solubility, solvation free energies, toxicity, partition coefficients, protein folding stability changes upon mutation, etc. 
    more » « less
  2. Skolnick, Jeffrey (Ed.)
    Systematically discovering protein-ligand interactions across the entire human and pathogen genomes is critical in chemical genomics, protein function prediction, drug discovery, and many other areas. However, more than 90% of gene families remain “dark”—i.e., their small-molecule ligands are undiscovered due to experimental limitations or human/historical biases. Existing computational approaches typically fail when the dark protein differs from those with known ligands. To address this challenge, we have developed a deep learning framework, called PortalCG, which consists of four novel components: (i) a 3-dimensional ligand binding site enhanced sequence pre-training strategy to encode the evolutionary links between ligand-binding sites across gene families; (ii) an end-to-end pretraining-fine-tuning strategy to reduce the impact of inaccuracy of predicted structures on function predictions by recognizing the sequence-structure-function paradigm; (iii) a new out-of-cluster meta-learning algorithm that extracts and accumulates information learned from predicting ligands of distinct gene families (meta-data) and applies the meta-data to a dark gene family; and (iv) a stress model selection step, using different gene families in the test data from those in the training and development data sets to facilitate model deployment in a real-world scenario. In extensive and rigorous benchmark experiments, PortalCG considerably outperformed state-of-the-art techniques of machine learning and protein-ligand docking when applied to dark gene families, and demonstrated its generalization power for target identifications and compound screenings under out-of-distribution (OOD) scenarios. Furthermore, in an external validation for the multi-target compound screening, the performance of PortalCG surpassed the rational design from medicinal chemists. Our results also suggest that a differentiable sequence-structure-function deep learning framework, where protein structural information serves as an intermediate layer, could be superior to conventional methodology where predicted protein structures were used for the compound screening. We applied PortalCG to two case studies to exemplify its potential in drug discovery: designing selective dual-antagonists of dopamine receptors for the treatment of opioid use disorder (OUD), and illuminating the understudied human genome for target diseases that do not yet have effective and safe therapeutics. Our results suggested that PortalCG is a viable solution to the OOD problem in exploring understudied regions of protein functional space. 
    more » « less
  3. G protein-coupled receptors (GPCRs) are the largest class of cell-surface receptor proteins with important functions in signal transduction and often serve as therapeutic drug targets. With the rapidly growing public data on three dimensional (3D) structures of GPCRs and GPCR-ligand interactions, computational prediction of GPCR ligand binding becomes a convincing option to high throughput screening and other experimental approaches during the beginning phases of ligand discovery. In this work, we set out to computationally uncover and understand the binding of a single ligand to GPCRs from several different families. Three-dimensional structural comparisons of the GPCRs that bind to the same ligand revealed local 3D structural similarities and often these regions overlap with locations of binding pockets. These pockets were found to be similar (based on backbone geometry and side-chain orientation using APoc), and they correlate positively with electrostatic properties of the pockets. Moreover, the more similar the pockets, the more likely a ligand binding to the pockets will interact with similar residues, have similar conformations, and produce similar binding affinities across the pockets. These findings can be exploited to improve protein function inference, drug repurposing and drug toxicity prediction, and accelerate the development of new drugs. 
    more » « less
  4. Abstract

    Amide−π interactions, in which an amide interacts with an aromatic group, are ubiquitous in biology, yet remain understudied relative to other noncovalent interactions. Recently, we demonstrated that an electrostatically tunable amide−π interaction is key to recognition of histone acyllysine by the AF9 YEATS domain, a reader protein which has emerged as a therapeutic target due to its dysregulation in cancer. Amide isosteres are commonly employed in drug discovery, often to prevent degradation by proteases, and have proven valuable in achieving selectivity when targeting epigenetic proteins. However, like amide−π interactions, interactions of amide isosteres with aromatic rings have not been thoroughly studied despite widespread use. Herein, we evaluate the recognition of a series of amide isosteres by the AF9 YEATS domain using genetic code expansion to evaluate the amide isostere−π interaction. We show that compared to the amide−π interaction with the native ligand, each isostere exhibits similar electrostatic tunability with an aromatic residue in the binding pocket, demonstrating that the isosteres maintain similar interactions with the aromatic residue. We identify a urea‐containing ligand that binds with enhanced affinity for the AF9 YEATS domain, offering a promising starting point for inhibitor development. Furthermore, we demonstrate that carbamate and urea isosteres of crotonyllysine are resistant to enzymatic removal by SIRT1, a protein that cleaves acyl post‐translational modifications, further indicating the potential of amide isosteres in YEATS domain inhibitor development. These results also provide experimental precedent for interactions of these common drug discovery moieties with aromatic rings that can inform computational methods.

     
    more » « less
  5. Accurate and efficient simulation of the thermodynamics and kinetics of protein–ligand interactions is crucial for computational drug discovery. Multiensemble Markov Model (MEMM) estimators can provide estimates of both binding rates and affinities from collections of short trajectories but have not been systematically explored for situations when a ligand is decoupled through scaling of non-bonded interactions. In this work, we compare the performance of two MEMM approaches for estimating ligand binding affinities and rates: (1) the transition-based reweighting analysis method (TRAM) and (2) a Maximum Caliber (MaxCal) based method. As a test system, we construct a small host–guest system where the ligand is a single uncharged Lennard-Jones (LJ) particle, and the receptor is an 11-particle icosahedral pocket made from the same atom type. To realistically mimic a protein–ligand binding system, the LJ ϵ parameter was tuned, and the system was placed in a periodic box with 860 TIP3P water molecules. A benchmark was performed using over 80 µs of unbiased simulation, and an 18-state Markov state model was used to estimate reference binding affinities and rates. We then tested the performance of TRAM and MaxCal when challenged with limited data. Both TRAM and MaxCal approaches perform better than conventional Markov state models, with TRAM showing better convergence and accuracy. We find that subsampling of trajectories to remove time correlation improves the accuracy of both TRAM and MaxCal and that in most cases, only a single biased ensemble to enhance sampled transitions is required to make accurate estimates.

     
    more » « less