Physics-based protein structure refinement in the era of artificial intelligence

Artificial Intelligence-Driven Computational Approaches in the Development of Anticancer Drugs

The integration of AI has revolutionized cancer drug development, transforming the landscape of drug discovery through sophisticated computational techniques. AI-powered models and algorithms have enhanced computer-aided drug design (CADD), offering unprecedented precision in identifying potential anticancer compounds. Traditionally, cancer drug design has been a complex, resource-intensive process, but AI introduces new opportunities to accelerate discovery, reduce costs, and optimize efficiency. This manuscript delves into the transformative applications of AI-driven methodologies in predicting and developing anticancer drugs, critically evaluating their potential to reshape the future of cancer therapeutics while addressing their challenges and limitations.

Predicting Protein-Ligand Binding Affinity Using Fusion Model of Spatial-Temporal Graph Neural Network and 3D Structure-Based Complex Graph

The investigation of molecular interactions between ligands and their target molecules is becoming more significant as protein structure data continues to develop. In this study, we introduce PLA-STGCNnet, a deep fusion spatial-temporal graph neural network designed to study protein-ligand interactions based on the 3D structural data of protein-ligand complexes. Unlike 1D protein sequences or 2D ligand graphs, the 3D graph representation offers a more precise portrayal of the complex interactions between proteins and ligands. Research studies have shown that our fusion model, PLA-STGCNnet, outperforms individual algorithms in accurately predicting binding affinity. The advantage of a fusion model is the ability to fully combine the advantages of multiple different models and improve overall performance by combining their features and outputs. Our fusion model shows satisfactory performance on different data sets, which proves its generalization ability and stability. The fusion-based model showed good performance in protein-ligand affinity prediction, and we successfully applied the model to drug screening. Our research underscores the promise of fusion spatial-temporal graph neural networks in addressing complex challenges in protein-ligand affinity prediction. The Python scripts for implementing various model components are accessible at https://github.com/ligaili01/PLA-STGCN.

Efficient Refinement of Complex Structures of Flexible Histone Peptides Using Post-Docking Molecular Dynamics Protocols

Histones are keys to many epigenetic events and their complexes have therapeutic and diagnostic importance. The determination of the structures of histone complexes is fundamental in the design of new drugs. Computational molecular docking is widely used for the prediction of target-ligand complexes. Large, linear peptides like the tail regions of histones are challenging ligands for docking due to their large conformational flexibility, extensive hydration, and weak interactions with the shallow binding pockets of their reader proteins. Thus, fast docking methods often fail to produce complex structures of such peptide ligands at a level appropriate for drug design. To address this challenge, and improve the structural quality of the docked complexes, post-docking refinement has been applied using various molecular dynamics (MD) approaches. However, a final consensus has not been reached on the desired MD refinement protocol. In this present study, MD refinement strategies were systematically explored on a set of problematic complexes of histone peptide ligands with relatively large errors in their docked geometries. Six protocols were compared that differ in their MD simulation parameters. In all cases, pre-MD hydration of the complex interface regions was applied to avoid the unwanted presence of empty cavities. The best-performing protocol achieved a median of 32% improvement over the docked structures in terms of the change in root mean squared deviations from the experimental references. The influence of structural factors and explicit hydration on the performance of post-docking MD refinements are also discussed to help with their implementation in future methods and applications.

Exploring the Alternative Conformation of a Known Protein Structure Based on Contact Map Prediction

The rapid development of deep learning-based methods has considerably advanced the field of protein structure prediction. The accuracy of predicting the 3D structures of simple proteins is comparable to that of experimentally determined structures, providing broad possibilities for structure-based biological studies. Another critical question is whether and how multistate structures can be predicted from a given protein sequence. In this study, analysis of tens of two-state proteins demonstrated that deep learning-based contact map predictions contain structural information on both states, which suggests that it is probably appropriate to change the target of deep learning-based protein structure prediction from one specific structure to multiple likely structures. Furthermore, by combining deep learning- and physics-based computational methods, we developed a protocol for exploring alternative conformations from a known structure of a given protein, by which we successfully approached the holo-state conformations of multiple representative proteins from their apo-state structures.

Improvement of protein tertiary and quaternary structure predictions using the ReFOLD refinement method and the AlphaFold2 recycling process

Motivation

The accuracy gap between predicted and experimental structures has been significantly reduced following the development of AlphaFold2 (AF2). However, for many targets, AF2 models still have room for improvement. In previous CASP experiments, highly computationally intensive MD simulation-based methods have been widely used to improve the accuracy of single 3D models. Here, our ReFOLD pipeline was adapted to refine AF2 predictions while maintaining high model accuracy at a modest computational cost. Furthermore, the AF2 recycling process was utilized to improve 3D models by using them as custom template inputs for tertiary and quaternary structure predictions.

Results

According to the Molprobity score, 94% of the generated 3D models by ReFOLD were improved. AF2 recycling showed an improvement rate of 87.5% (using MSAs) and 81.25% (using single sequences) for monomeric AF2 models and 100% (MSA) and 97.8% (single sequence) for monomeric non-AF2 models, as measured by the average change in lDDT. By the same measure, the recycling of multimeric models showed an improvement rate of as much as 80% for AF2-Multimer (AF2M) models and 94% for non-AF2M models.

Availability and implementation

Refinement using AlphaFold2-Multimer recycling is available as part of the MultiFOLD docker package (https://hub.docker.com/r/mcguffin/multifold). The ReFOLD server is available at https://www.reading.ac.uk/bioinf/ReFOLD/ and the modified scripts can be downloaded from https://www.reading.ac.uk/bioinf/downloads/.

Supplementary information

Supplementary data are available at Bioinformatics Advances online.

Machine Learning-Based Modeling of Olfactory Receptors in Their Inactive State: Human OR51E2 as a Case Study

Atomistic-level investigation of olfactory receptors (ORs) is a challenging task due to the experimental/computational difficulties in the structural determination/prediction for members of this family of G-protein coupled receptors. Here, we have developed a protocol that performs a series of molecular dynamics simulations from a set of structures predicted de novo by recent machine learning algorithms and apply it to a well-studied receptor, the human OR51E2. Our study demonstrates the need for simulations to refine and validate such models. Furthermore, we demonstrate the need for the sodium ion at a binding site near D^2.50 and E^3.39 to stabilize the inactive state of the receptor. Considering the conservation of these two acidic residues across human ORs, we surmise this requirement also applies to the other ∼400 members of this family. Given the almost concurrent publication of a CryoEM structure of the same receptor in the active state, we propose this protocol as an in silico complement to the growing field of ORs structure determination.

Atomic protein structure refinement using all-atom graph representations and SE(3)-equivariant graph transformer

MOTIVATION

The state-of-art protein structure prediction methods such as AlphaFold are being widely used to predict structures of uncharacterized proteins in biomedical research. There is a significant need to further improve the quality and nativeness of the predicted structures to enhance their usability. In this work, we develop ATOMRefine, a deep learning-based, end-to-end, all-atom protein structural model refinement method. It uses a SE(3)-equivariant graph transformer network to directly refine protein atomic coordinates in a predicted tertiary structure represented as a molecular graph.

RESULTS

The method is first trained and tested on the structural models in AlphaFoldDB whose experimental structures are known, and then blindly tested on 69 CASP14 regular targets and 7 CASP14 refinement targets. ATOMRefine improves the quality of both backbone atoms and all-atom conformation of the initial structural models generated by AlphaFold. It also performs better than two state-of-the-art refinement methods in multiple evaluation metrics including an all-atom model quality score-the MolProbity score based on the analysis of all-atom contacts, bond length, atom clashes, torsion angles, and side-chain rotamers. As ATOMRefine can refine a protein structure quickly, it provides a viable, fast solution for improving protein geometry and fixing structural errors of predicted structures through direct coordinate refinement.

AVAILABILITY AND IMPLEMENTATION

The source code of ATOMRefine is available in the GitHub repository (https://github.com/BioinfoMachineLearning/ATOMRefine). All the required data for training and testing are available at https://doi.org/10.5281/zenodo.6944368.

Protein Function Analysis through Machine Learning

Machine learning (ML) has been an important arsenal in computational biology used to elucidate protein function for decades. With the recent burgeoning of novel ML methods and applications, new ML approaches have been incorporated into many areas of computational biology dealing with protein function. We examine how ML has been integrated into a wide range of computational models to improve prediction accuracy and gain a better understanding of protein function. The applications discussed are protein structure prediction, protein engineering using sequence modifications to achieve stability and druggability characteristics, molecular docking in terms of protein-ligand binding, including allosteric effects, protein-protein interactions and protein-centric drug discovery. To quantify the mechanisms underlying protein function, a holistic approach that takes structure, flexibility, stability, and dynamics into account is required, as these aspects become inseparable through their interdependence. Another key component of protein function is conformational dynamics, which often manifest as protein kinetics. Computational methods that use ML to generate representative conformational ensembles and quantify differences in conformational ensembles important for function are included in this review. Future opportunities are highlighted for each of these topics.

A-Prot: protein structure modeling using MSA transformer

Background

The accuracy of protein 3D structure prediction has been dramatically improved with the help of advances in deep learning. In the recent CASP14, Deepmind demonstrated that their new version of AlphaFold (AF) produces highly accurate 3D models almost close to experimental structures. The success of AF shows that the multiple sequence alignment of a sequence contains rich evolutionary information, leading to accurate 3D models. Despite the success of AF, only the prediction code is open, and training a similar model requires a vast amount of computational resources. Thus, developing a lighter prediction model is still necessary.

Results

In this study, we propose a new protein 3D structure modeling method, A-Prot, using MSA Transformer, one of the state-of-the-art protein language models. An MSA feature tensor and row attention maps are extracted and converted into 2D residue-residue distance and dihedral angle predictions for a given MSA. We demonstrated that A-Prot predicts long-range contacts better than the existing methods. Additionally, we modeled the 3D structures of the free modeling and hard template-based modeling targets of CASP14. The assessment shows that the A-Prot models are more accurate than most top server groups of CASP14.

Conclusion

These results imply that A-Prot accurately captures the evolutionary and structural information of proteins with relatively low computational cost. Thus, A-Prot can provide a clue for the development of other protein property prediction methods.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-022-04628-8.

Ins and outs of AlphaFold2 transmembrane protein structure predictions

Transmembrane (TM) proteins are major drug targets, but their structure determination, a prerequisite for rational drug design, remains challenging. Recently, the DeepMind's AlphaFold2 machine learning method greatly expanded the structural coverage of sequences with high accuracy. Since the employed algorithm did not take specific properties of TM proteins into account, the reliability of the generated TM structures should be assessed. Therefore, we quantitatively investigated the quality of structures at genome scales, at the level of ABC protein superfamily folds and for specific membrane proteins (e.g. dimer modeling and stability in molecular dynamics simulations). We tested template-free structure prediction with a challenging TM CASP14 target and several TM protein structures published after AlphaFold2 training. Our results suggest that AlphaFold2 performs well in the case of TM proteins and its neural network is not overfitted. We conclude that cautious applications of AlphaFold2 structural models will advance TM protein-associated studies at an unexpected level.

Probing the Formation, Structure and Free Energy Relationships of M Protein Dimers of SARS-CoV-2

The M protein of the novel coronavirus 2019 (SARS-CoV-2) is the major structural component of the viral envelope and is also the minimum requirement for virus particle budding. M proteins generally exist as dimers. In virus assembly, they are the main driving force for envelope formation through lateral interactions and interactions with other viral structural proteins that play a central role. We built 100 candidate models and finally analyzed the six most convincing structural features of the SARS-CoV-2 M protein dimer based on long-timescale molecular dynamics (MD) simulations, multiple free energy analyses (potential mean force (PMF) and molecular mechanics Poisson-Boltzmann surface area (MMPBSA)) and principal component analysis (PCA) to obtain the most reasonable structure. The dimer stability was found to depend on the Leu-Ile zipper motif and aromatic amino acids in the transmembrane domain (TMD). Furthermore, the C-terminal domain (CTD) effects were relatively small. These results highlight a model in which there is sufficient binding affinity between the TMDs of M proteins to form dimers through the residues at the interface of the three transmembrane helices (TMHs). This study aims to help find more effective inhibitors of SARS-CoV-2 M dimers and to develop vaccines based on structural information.

Evaluation of model refinement in CASP14

We report here an assessment of the model refinement category of the 14th round of Critical Assessment of Structure Prediction (CASP14). As before, predictors submitted up to five ranked refinements, along with associated residue-level error estimates, for targets that had a wide range of starting quality. The ability of groups to accurately rank their submissions and to predict coordinate error varied widely. Overall, only four groups out-performed a "naïve predictor" corresponding to the resubmission of the starting model. Among the top groups, there are interesting differences of approach and in the spread of improvements seen: some methods are more conservative, others more adventurous. Some targets were "double-barreled" for which predictors were offered a high-quality AlphaFold 2 (AF2)-derived prediction alongside another of lower quality. The AF2-derived models were largely unimprovable, many of their apparent errors being found to reside at domain and, especially, crystal lattice contacts. Refinement is shown to have a mixed impact overall on structure-based function annotation methods to predict nucleic acid binding, spot catalytic sites, and dock protein structures.

Assessing the utility of CASP14 models for molecular replacement

The assessment of CASP models for utility in molecular replacement is a measure of their use in a valuable real‐world application. In CASP7, the metric for molecular replacement assessment involved full likelihood‐based molecular replacement searches; however, this restricted the assessable targets to crystal structures with only one copy of the target in the asymmetric unit, and to those where the search found the correct pose. In CASP10, full molecular replacement searches were replaced by likelihood‐based rigid‐body refinement of models superimposed on the target using the LGA algorithm, with the metric being the refined log‐likelihood‐gain (LLG) score. This enabled multi‐copy targets and very poor models to be evaluated, but a significant further issue remained: the requirement of diffraction data for assessment. We introduce here the relative‐expected‐LLG (reLLG), which is independent of diffraction data. This reLLG is also independent of any crystal form, and can be calculated regardless of the source of the target, be it X‐ray, NMR or cryo‐EM. We calibrate the reLLG against the LLG for targets in CASP14, showing that it is a robust measure of both model and group ranking. Like the LLG, the reLLG shows that accurate coordinate error estimates add substantial value to predicted models. We find that refinement by CASP groups can often convert an inadequate initial model into a successful MR search model. Consistent with findings from others, we show that the AlphaFold2 models are sufficiently good, and reliably so, to surpass other current model generation strategies for attempting molecular replacement phasing.