Evaluation of AlphaFold2 structures as docking targets

AlphaFold2 is a promising new tool for researchers to predict protein structures and generate high-quality models, with low backbone and global root-mean-square deviation (RMSD) when compared with experimental structures. However, it is unclear if the structures predicted by AlphaFold2 will be valuable targets of docking. To address this question, we redocked ligands in the PDBbind datasets against the experimental co-crystallized receptor structures and against the AlphaFold2 structures using AutoDock-GPU. We find that the quality measure provided during structure prediction is not a good predictor of docking performance, despite accurately reflecting the quality of the alpha carbon alignment with experimental structures. Removing low-confidence regions of the predicted structure and making side chains flexible improves the docking outcomes. Overall, despite high-quality prediction of backbone conformation, fine structural details limit the naive application of AlphaFold2 models as docking targets.     

The structural coverage of the human proteome before and after AlphaFold

The protein structure field is experiencing a revolution. From the increased throughput of techniques to determine experimental structures, to developments such as cryo-EM that allow us to find the structures of large protein complexes or, more recently, the development of artificial intelligence tools, such as AlphaFold, that can predict with high accuracy the folding of proteins for which the availability of homology templates is limited. Here we quantify the effect of the recently released AlphaFold database of protein structural models in our knowledge on human proteins. Our results indicate that our current baseline for structural coverage of 48%, considering experimentally-derived or template-based homology models, elevates up to 76% when including AlphaFold predictions. At the same time the fraction of dark proteome is reduced from 26% to just 10% when AlphaFold models are considered. Furthermore, although the coverage of disease-associated genes and mutations was near complete before AlphaFold release (69% of Clinvar pathogenic mutations and 88% of oncogenic mutations), AlphaFold models still provide an additional coverage of 3% to 13% of these critically important sets of biomedical genes and mutations. Finally, we show how the contribution of AlphaFold models to the structural coverage of non-human organisms, including important pathogenic bacteria, is significantly larger than that of the human proteome. Overall, our results show that the sequence-structure gap of human proteins has almost disappeared, an outstanding success of direct consequences for the knowledge on the human genome and the derived medical applications.     

Identification of Iron-Sulfur (Fe-S) Cluster and Zinc (Zn) Binding Sites Within Proteomes Predicted by DeepMind’s AlphaFold2 Program Dramatically Expands the Metalloproteome

DeepMind’s AlphaFold2 software has ushered in a revolution in high quality, 3D protein structure prediction. In very recent work by the DeepMind team, structure predictions have been made for entire proteomes of twenty-one organisms, with >360,000 structures made available for download. Here we show that thousands of novel binding sites for iron-sulfur (Fe-S) clusters and zinc (Zn) ions can be identified within these predicted structures by exhaustive enumeration of all potential ligand-binding orientations. We demonstrate that AlphaFold2 routinely makes highly specific predictions of ligand binding sites: for example, binding sites that are comprised exclusively of four cysteine sidechains fall into three clusters, representing binding sites for 4Fe-4S clusters, 2Fe-2S clusters, or individual Zn ions. We show further: (a) that the majority of known Fe-S cluster and Zn binding sites documented in UniProt are recovered by the AlphaFold2 structures, (b) that there are occasional disputes between AlphaFold2 and UniProt with AlphaFold2 predicting highly plausible alternative binding sites, (c) that the Fe-S cluster binding sites that we identify in E. coli agree well with previous bioinformatics predictions, (d) that cysteines predicted here to be part of ligand binding sites show little overlap with those shown via chemoproteomics techniques to be highly reactive, and (e) that AlphaFold2 occasionally appears to build erroneous disulfide bonds between cysteines that should instead coordinate a ligand. These results suggest that AlphaFold2 could be an important tool for the functional annotation of proteomes, and the methodology presented here is likely to be useful for predicting other ligand-binding sites.     

NMR hawk‐eyed view of AlphaFold2 structures

The prediction of the three‐dimensional (3D) structure of proteins from the amino acid sequence made a stunning breakthrough reaching atomic accuracy. Using the neural network‐based method AlphaFold2, 3D structures of almost the entire human proteome have been predicted and made available (https://www.alphafold.ebi.ac.uk). To gain insight into how well AlphaFold2 structures represent the conformation of proteins in solution, I here compare the AlphaFold2 structures of selected small proteins with their 3D structures that were determined by nuclear magnetic resonance (NMR) spectroscopy. Proteins were selected for which the 3D solution structures were determined on the basis of a very large number of distance restraints and residual dipolar couplings and are thus some of the best‐resolved solution structures of proteins to date. The quality of the backbone conformation of the AlphaFold2 structures is assessed by fitting a large set of experimental residual dipolar couplings (RDCs). The analysis shows that experimental RDCs fit extremely well to the AlphaFold2 structures predicted for GB3, DinI, and ubiquitin. In the case of GB3, the accuracy of the AlphaFold2 structure even surpasses that of a 1.1 Å crystal structure. Fitting of experimental RDCs furthermore allows identification of AlphaFold2 structures that are best representative of the protein''s conformation in solution as seen for the EF hands of the N‐terminal domain of Ca²⁺‐ligated calmodulin. Taken together, the analysis shows that structures predicted by AlphaFold2 can be highly representative of the solution conformation of proteins. The combination of AlphaFold2 structures with RDCs promises to be a powerful approach to study structural changes in proteins.     

Prediction of protein oligomer structures using GALAXY in CASP13

Many proteins need to form oligomers to be functional, so oligomer structures provide important clues to biological roles of proteins. Prediction of oligomer structures therefore can be a useful tool in the absence of experimentally resolved structures. In this article, we describe the server and human methods that we used to predict oligomer structures in the CASP13 experiment. Performances of the methods on the 42 CASP13 oligomer targets consisting of 30 homo-oligomers and 12 hetero-oligomers are discussed. Our server method, Seok-assembly, generated models with interface contact similarity measure greater than 0.2 as model 1 for 11 homo-oligomer targets when proper templates existed in the database. Model refinement methods such as loop modeling and molecular dynamics (MD)-based overall refinement failed to improve model qualities when target proteins have domains not covered by templates or when chains have very small interfaces. In human predictions, additional experimental data such as low-resolution electron microscopy (EM) map were utilized. EM data could assist oligomer structure prediction by providing a global shape of the complex structure.     

Integrative Structural Biology in the Era of Accurate Structure Prediction

Characterizing the three-dimensional structure of macromolecules is central to understanding their function. Traditionally, structures of proteins and their complexes have been determined using experimental techniques such as X-ray crystallography, NMR, or cryo-electron microscopy—applied individually or in an integrative manner. Meanwhile, however, computational methods for protein structure prediction have been improving their accuracy, gradually, then suddenly, with the breakthrough advance by AlphaFold2, whose models of monomeric proteins are often as accurate as experimental structures. This breakthrough foreshadows a new era of computational methods that can build accurate models for most monomeric proteins. Here, we envision how such accurate modeling methods can combine with experimental structural biology techniques, enhancing integrative structural biology. We highlight the challenges that arise when considering multiple structural conformations, protein complexes, and polymorphic assemblies. These challenges will motivate further developments, both in modeling programs and in methods to solve experimental structures, towards better and quicker investigation of structure–function relationships.     

Using a library of structural templates to recognise catalytic sites and explore their evolution in homologous families

Catalytic site structure is normally highly conserved between distantly related enzymes. As a consequence, templates representing catalytic sites have the potential to succeed at function prediction in cases where methods based on sequence or overall structure fail. There are many methods for searching protein structures for matches to structural templates, but few validated template libraries to use with these methods. We present a library of structural templates representing catalytic sites, based on information from the scientific literature. Furthermore, we analyse homologous template families to discover the diversity within families and the utility of templates for active site recognition. Templates representing the catalytic sites of homologous proteins mostly differ by less than 1A root mean square deviation, even when the sequence similarity between the two proteins is low. Within these sets of homologues there is usually no discernible relationship between catalytic site structure similarity and sequence similarity. Because of this structural conservation of catalytic sites, the templates can discriminate between matches to related proteins and random matches with over 85% sensitivity and predictive accuracy. Templates based on protein backbone positions are more discriminating than those based on side-chain atoms. These analyses show encouraging prospects for prediction of functional sites in structural genomics structures of unknown function, and will be of use in analyses of convergent evolution and exploring relationships between active site geometry and chemistry. The template library can be queried via a web server at and is available for download.     

Beyond the Twilight Zone: Automated prediction of structural properties of proteins by recursive neural networks and remote homology information

The prediction of 1D structural properties of proteins is an important step toward the prediction of protein structure and function, not only in the ab initio case but also when homology information to known structures is available. Despite this the vast majority of 1D predictors do not incorporate homology information into the prediction process. We develop a novel structural alignment method, SAMD, which we use to build alignments of putative remote homologues that we compress into templates of structural frequency profiles. We use these templates as additional input to ensembles of recursive neural networks, which we specialise for the prediction of query sequences that show only remote homology to any Protein Data Bank structure. We predict four 1D structural properties – secondary structure, relative solvent accessibility, backbone structural motifs, and contact density. Secondary structure prediction accuracy, tested by five‐fold cross‐validation on a large set of proteins allowing less than 25% sequence identity between training and test set and query sequences and templates, exceeds 82%, outperforming its ab initio counterpart, other state‐of‐the‐art secondary structure predictors (Jpred 3 and PSIPRED) and two other systems based on PSI‐BLAST and COMPASS templates. We show that structural information from homologues improves prediction accuracy well beyond the Twilight Zone of sequence similarity, even below 5% sequence identity, for all four structural properties. Significant improvement over the extraction of structural information directly from PDB templates suggests that the combination of sequence and template information is more informative than templates alone. Proteins 2009. © 2009 Wiley‐Liss, Inc.     

Protein function prediction using local 3D templates

The prediction of a protein's function from its 3D structure is becoming more and more important as the worldwide structural genomics initiatives gather pace and continue to solve 3D structures, many of which are of proteins of unknown function. Here, we present a methodology for predicting function from structure that shows great promise. It is based on 3D templates that are defined as specific 3D conformations of small numbers of residues. We use four types of template, covering enzyme active sites, ligand-binding residues, DNA-binding residues and reverse templates. The latter are templates generated from the target structure itself and scanned against a representative subset of all known protein structures. Together, the templates provide a fairly thorough coverage of the known structures and ensure that if there is a match to a known structure it is unlikely to be missed. A new scoring scheme provides a highly sensitive means of discriminating between true positive and false positive template matches. In all, the methodology provides a powerful new tool for function prediction to complement those already in use.     

Comparative Sequence and Structural Analyses of G-Protein-Coupled Receptor Crystal Structures and Implications for Molecular Models

Background

Up until recently the only available experimental (high resolution) structure of a G-protein-coupled receptor (GPCR) was that of bovine rhodopsin. In the past few years the determination of GPCR structures has accelerated with three new receptors, as well as squid rhodopsin, being successfully crystallized. All share a common molecular architecture of seven transmembrane helices and can therefore serve as templates for building molecular models of homologous GPCRs. However, despite the common general architecture of these structures key differences do exist between them. The choice of which experimental GPCR structure(s) to use for building a comparative model of a particular GPCR is unclear and without detailed structural and sequence analyses, could be arbitrary. The aim of this study is therefore to perform a systematic and detailed analysis of sequence-structure relationships of known GPCR structures.

Methodology

We analyzed in detail conserved and unique sequence motifs and structural features in experimentally-determined GPCR structures. Deeper insight into specific and important structural features of GPCRs as well as valuable information for template selection has been gained. Using key features a workflow has been formulated for identifying the most appropriate template(s) for building homology models of GPCRs of unknown structure. This workflow was applied to a set of 14 human family A GPCRs suggesting for each the most appropriate template(s) for building a comparative molecular model.

Conclusions

The available crystal structures represent only a subset of all possible structural variation in family A GPCRs. Some GPCRs have structural features that are distributed over different crystal structures or which are not present in the templates suggesting that homology models should be built using multiple templates. This study provides a systematic analysis of GPCR crystal structures and a consistent method for identifying suitable templates for GPCR homology modelling that will help to produce more reliable three-dimensional models.     

Ligand binding and homology modelling of insect odorant‐binding proteins

This review describes the main characteristics of odorant‐binding proteins (OBPs) for homology modelling and presents a summary of structure prediction studies on insect OBPs, along with the steps involved and some limitations and improvements. The technique involves a computing approach to model protein structures and is based on a comparison between a target (unknown structure) and one or more templates (experimentally determined structures). As targets for structure prediction, OBPs are considered to play a functional role for recognition, desorption, scavenging, protection and transportation of hydrophobic molecules (odourants) across an aqueous environment (lymph) to olfactory receptor neurones (ORNs) located in sensilla, the main olfactory units of insect antennae. Lepidopteran pheromone‐binding proteins, a subgroup of OBPs, are characterized by remarkable structural features, in which high sequence identities (approximately 30%) among these OBPs and a large number of available templates can facilitate the prediction of precise homology models. Approximately 30 studies have been performed on insect OBPs using homology modelling as a tool to predict their structures. Although some of the studies have assessed ligand‐binding affinity using structural information and biochemical measurements, few have performed docking and molecular dynamic (MD) simulations as a virtual method to predict best ligands. Docking and MD simulations are discussed in the context of discovery of novel semiochemicals (super‐ligands) using homology modelling to conceive further strategies in insect management.     

Computational modeling of membrane proteins

The determination of membrane protein (MP) structures has always trailed that of soluble proteins due to difficulties in their overexpression, reconstitution into membrane mimetics, and subsequent structure determination. The percentage of MP structures in the protein databank (PDB) has been at a constant 1–2% for the last decade. In contrast, over half of all drugs target MPs, only highlighting how little we understand about drug‐specific effects in the human body. To reduce this gap, researchers have attempted to predict structural features of MPs even before the first structure was experimentally elucidated. In this review, we present current computational methods to predict MP structure, starting with secondary structure prediction, prediction of trans‐membrane spans, and topology. Even though these methods generate reliable predictions, challenges such as predicting kinks or precise beginnings and ends of secondary structure elements are still waiting to be addressed. We describe recent developments in the prediction of 3D structures of both α‐helical MPs as well as β‐barrels using comparative modeling techniques, de novo methods, and molecular dynamics (MD) simulations. The increase of MP structures has (1) facilitated comparative modeling due to availability of more and better templates, and (2) improved the statistics for knowledge‐based scoring functions. Moreover, de novo methods have benefited from the use of correlated mutations as restraints. Finally, we outline current advances that will likely shape the field in the forthcoming decade. Proteins 2015; 83:1–24. © 2014 Wiley Periodicals, Inc.     

PFRES: protein fold classification by using evolutionary information and predicted secondary structure

MOTIVATION: The number of protein families has been estimated to be as small as 1000. Recent study shows that the growth in discovery of novel structures that are deposited into PDB and the related rate of increase of SCOP categories are slowing down. This indicates that the protein structure space will be soon covered and thus we may be able to derive most of remaining structures by using the known folding patterns. Present tertiary structure prediction methods behave well when a homologous structure is predicted, but give poorer results when no homologous templates are available. At the same time, some proteins that share twilight-zone sequence identity can form similar folds. Therefore, determination of structural similarity without sequence similarity would be beneficial for prediction of tertiary structures. RESULTS: The proposed PFRES method for automated protein fold classification from low identity (<35%) sequences obtains 66.4% and 68.4% accuracy for two test sets, respectively. PFRES obtains 6.3-12.4% higher accuracy than the existing methods. The prediction accuracy of PFRES is shown to be statistically significantly better than the accuracy of competing methods. Our method adopts a carefully designed, ensemble-based classifier, and a novel, compact and custom-designed feature representation that includes nearly 90% less features than the representation of the most accurate competing method (36 versus 283). The proposed representation combines evolutionary information by using the PSI-BLAST profile-based composition vector and information extracted from the secondary structure predicted with PSI-PRED. AVAILABILITY: The method is freely available from the authors upon request.     

Capturing the geometry,function, and evolution of enzymes with 3D templates

Structural templates are 3D signatures representing protein functional sites, such as ligand binding cavities, metal coordination motifs, or catalytic sites. Here we explore methods to generate template libraries and algorithms to query structures for conserved 3D motifs. Applications of templates are discussed, as well as some exemplar cases for examining evolutionary links in enzymes. We also introduce the concept of using more than one template per structure to represent flexible sites, as an approach to better understand catalysis through snapshots captured in enzyme structures. Functional annotation from structure is an important topic that has recently resurfaced due to the new more accurate methods of protein structure prediction. Therefore, we anticipate that template‐based functional site detection will be a powerful tool in the task of characterizing a vast number of new protein models.     

Internal organization of large protein families: relationship between the sequence, structure, and function-based clustering

The protein universe can be organized in families that group proteins sharing common ancestry. Such families display variable levels of structural and functional divergence, from homogenous families, where all members have the same function and very similar structure, to very divergent families, where large variations in function and structure are observed. For practical purposes of structure and function prediction, it would be beneficial to identify sub-groups of proteins with highly similar structures (iso-structural) and/or functions (iso-functional) within divergent protein families. We compared three algorithms in their ability to cluster large protein families and discuss whether any of these methods could reliably identify such iso-structural or iso-functional groups. We show that clustering using profile-sequence and profile-profile comparison methods closely reproduces clusters based on similarities between 3D structures or clusters of proteins with similar biological functions. In contrast, the still commonly used sequence-based methods with fixed thresholds result in vast overestimates of structural and functional diversity in protein families. As a result, these methods also overestimate the number of protein structures that have to be determined to fully characterize structural space of such families. The fact that one can build reliable models based on apparently distantly related templates is crucial for extracting maximal amount of information from new sequencing projects.     

Modeling conformational states of proteins with AlphaFold

Many proteins exert their function by switching among different structures. Knowing the conformational ensembles affiliated with these states is critical to elucidate key mechanistic aspects that govern protein function. While experimental determination efforts are still bottlenecked by cost, time, and technical challenges, the machine-learning technology AlphaFold showed near experimental accuracy in predicting the three-dimensional structure of monomeric proteins. However, an AlphaFold ensemble of models usually represents a single conformational state with minimal structural heterogeneity. Consequently, several pipelines have been proposed to either expand the structural breadth of an ensemble or bias the prediction toward a desired conformational state. Here, we analyze how those pipelines work, what they can and cannot predict, and future directions.     

Improving protein tertiary structure prediction by deep learning and distance prediction in CASP14

Substantial progresses in protein structure prediction have been made by utilizing deep-learning and residue-residue distance prediction since CASP13. Inspired by the advances, we improve our CASP14 MULTICOM protein structure prediction system by incorporating three new components: (a) a new deep learning-based protein inter-residue distance predictor to improve template-free (ab initio) tertiary structure prediction, (b) an enhanced template-based tertiary structure prediction method, and (c) distance-based model quality assessment methods empowered by deep learning. In the 2020 CASP14 experiment, MULTICOM predictor was ranked seventh out of 146 predictors in tertiary structure prediction and ranked third out of 136 predictors in inter-domain structure prediction. The results demonstrate that the template-free modeling based on deep learning and residue-residue distance prediction can predict the correct topology for almost all template-based modeling targets and a majority of hard targets (template-free targets or targets whose templates cannot be recognized), which is a significant improvement over the CASP13 MULTICOM predictor. Moreover, the template-free modeling performs better than the template-based modeling on not only hard targets but also the targets that have homologous templates. The performance of the template-free modeling largely depends on the accuracy of distance prediction closely related to the quality of multiple sequence alignments. The structural model quality assessment works well on targets for which enough good models can be predicted, but it may perform poorly when only a few good models are predicted for a hard target and the distribution of model quality scores is highly skewed. MULTICOM is available at https://github.com/jianlin-cheng/MULTICOM_Human_CASP14/tree/CASP14_DeepRank3 and https://github.com/multicom-toolbox/multicom/tree/multicom_v2.0 .     

Complementing machine learning‐based structure predictions with native mass spectrometry

The advent of machine learning‐based structure prediction algorithms such as AlphaFold2 (AF2) and RoseTTa Fold have moved the generation of accurate structural models for the entire cellular protein machinery into the reach of the scientific community. However, structure predictions of protein complexes are based on user‐provided input and may require experimental validation. Mass spectrometry (MS) is a versatile, time‐effective tool that provides information on post‐translational modifications, ligand interactions, conformational changes, and higher‐order oligomerization. Using three protein systems, we show that native MS experiments can uncover structural features of ligand interactions, homology models, and point mutations that are undetectable by AF2 alone. We conclude that machine learning can be complemented with MS to yield more accurate structural models on a small and large scale.     

AlphaFold and Implications for Intrinsically Disordered Proteins

Accurate predictions of the three-dimensional structures of proteins from their amino acid sequences have come of age. AlphaFold, a deep learning-based approach to protein structure prediction, shows remarkable success in independent assessments of prediction accuracy. A significant epoch in structural bioinformatics was the structural annotation of over 98% of protein sequences in the human proteome. Interestingly, many predictions feature regions of very low confidence, and these regions largely overlap with intrinsically disordered regions (IDRs). That over 30% of regions within the proteome are disordered is congruent with estimates that have been made over the past two decades, as intense efforts have been undertaken to generalize the structure–function paradigm to include the importance of conformational heterogeneity and dynamics. With structural annotations from AlphaFold in hand, there is the temptation to draw inferences regarding the “structures” of IDRs and their interactomes. Here, we offer a cautionary note regarding the misinterpretations that might ensue and highlight efforts that provide concrete understanding of sequence-ensemble-function relationships of IDRs. This perspective is intended to emphasize the importance of IDRs in sequence-function relationships (SERs) and to highlight how one might go about extracting quantitative SERs to make sense of how IDRs function.