ResPRE: high-accuracy protein contact prediction by coupling precision matrix with deep residual neural networks

MolLoG: A Molecular Level Interpretability Model Bridging Local to Global for Predicting Drug Target Interactions

Developing new pharmaceuticals is a costly and time-consuming endeavor fraught with significant safety risks. A critical aspect of drug research and disease therapy is discerning the existence of interactions between drugs and proteins. The evolution of deep learning (DL) in computer science has been remarkably aided in this regard in recent years. Yet, two challenges remain: (i) balancing the extraction of profound, local cohesive characteristics while warding off gradient disappearance and (ii) globally representing and understanding the interactions between the drug and target local attributes, which is vital for delivering molecular level insights indispensable to drug development. In response to these challenges, we propose a DL network structure, MolLoG, primarily comprising two modules: local feature encoders (LFE) and global interactive learning (GIL). Within the LFE module, graph convolution networks and leap blocks capture the local features of drug and protein molecules, respectively. The GIL module enables the efficient amalgamation of feature information, facilitating the global learning of feature structural semantics and procuring multihead attention weights for abstract features stemming from two modalities, providing biologically pertinent explanations for black-box results. Finally, predictive outcomes are achieved by decoding the unified representation via a multilayer perceptron. Our experimental analysis reveals that MolLoG outperforms several cutting-edge baselines across four data sets, delivering superior overall performance and providing satisfactory results when elucidating various facets of drug-target interaction predictions.

Can Protein Structure Prediction Methods Capture Alternative Conformations of Membrane Transporters?

Understanding the conformational dynamics of proteins, such as the inward-facing (IF) and outward-facing (OF) transition observed in transporters, is vital for elucidating their functional mechanisms. Despite significant advances in protein structure prediction (PSP) over the past three decades, most efforts have been focused on single-state prediction, leaving multistate or alternative conformation prediction (ACP) relatively unexplored. This discrepancy has led to the development of highly accurate PSP methods such as AlphaFold, yet their capabilities for ACP remain limited. To investigate the performance of current PSP methods in ACP, we curated a data set, named IOMemP, consisting of 32 experimentally determined high-resolution IF and OF structures of 16 membrane proteins with substantial conformational changes. We benchmarked 12 representative PSP methods, along with two recent multistate methods based on AlphaFold, against this data set. Our findings reveal a remarkably consistent preference for specific states across various PSP methods. We elucidated how coevolution information in MSAs influences state preference. Moreover, we showed that AlphaFold, when excluding coevolution information, estimated similar energies between the experimental IF and OF conformations, indicating that the energy model learned by AlphaFold is not biased toward any particular state. Our IOMemP data set and benchmark results are anticipated to advance the development of robust ACP methods.

Protein language model-embedded geometric graphs power inter-protein contact prediction

Accurate prediction of contacting residue pairs between interacting proteins is very useful for structural characterization of protein-protein interactions. Although significant improvement has been made in inter-protein contact prediction recently, there is still a large room for improving the prediction accuracy. Here we present a new deep learning method referred to as PLMGraph-Inter for inter-protein contact prediction. Specifically, we employ rotationally and translationally invariant geometric graphs obtained from structures of interacting proteins to integrate multiple protein language models, which are successively transformed by graph encoders formed by geometric vector perceptrons and residual networks formed by dimensional hybrid residual blocks to predict inter-protein contacts. Extensive evaluation on multiple test sets illustrates that PLMGraph-Inter outperforms five top inter-protein contact prediction methods, including DeepHomo, GLINTER, CDPred, DeepHomo2, and DRN-1D2D_Inter, by large margins. In addition, we also show that the prediction of PLMGraph-Inter can complement the result of AlphaFold-Multimer. Finally, we show leveraging the contacts predicted by PLMGraph-Inter as constraints for protein-protein docking can dramatically improve its performance for protein complex structure prediction.

Recent Progress of Protein Tertiary Structure Prediction

The prediction of three-dimensional (3D) protein structure from amino acid sequences has stood as a significant challenge in computational and structural bioinformatics for decades. Recently, the widespread integration of artificial intelligence (AI) algorithms has substantially expedited advancements in protein structure prediction, yielding numerous significant milestones. In particular, the end-to-end deep learning method AlphaFold2 has facilitated the rise of structure prediction performance to new heights, regularly competitive with experimental structures in the 14th Critical Assessment of Protein Structure Prediction (CASP14). To provide a comprehensive understanding and guide future research in the field of protein structure prediction for researchers, this review describes various methodologies, assessments, and databases in protein structure prediction, including traditionally used protein structure prediction methods, such as template-based modeling (TBM) and template-free modeling (FM) approaches; recently developed deep learning-based methods, such as contact/distance-guided methods, end-to-end folding methods, and protein language model (PLM)-based methods; multi-domain protein structure prediction methods; the CASP experiments and related assessments; and the recently released AlphaFold Protein Structure Database (AlphaFold DB). We discuss their advantages, disadvantages, and application scopes, aiming to provide researchers with insights through which to understand the limitations, contexts, and effective selections of protein structure prediction methods in protein-related fields.

AttCON: With better MSAs and attention mechanism for accurate protein contact map prediction

Protein contact map prediction is a critical and vital step in protein structure prediction, and its accuracy is highly contingent upon the feature representations of protein sequence information and the efficacy of deep learning models. In this paper, we propose an algorithm, DeepMSA+, to generate protein multiple sequence alignments (MSAs) and to construct feature representations based on co-evolutionary information and sequence information derived from MSAs. We also propose an improved deep learning model, AttCON, for training input features to predict protein contact map. The model incorporates an attention module, and by comparing different attention modules, we find a parameter-free attention module suitable for contact map prediction. Additionally, we use the Focal Loss function to better address the data imbalance issue in protein contact map. We also developed a weighted evaluation index (W score) for model evaluation, which takes into account a wide range of metrics. W score is comprehensive in its scope, with a particular focus on the precision of predictions for medium-range and long-range contacts. Experimental results show that AttCON achieves good precision results on datasets from CASP11 to CASP15. Compared to some state-of-the-art methods, it achieves an average improvement of over 5% in both medium-range and long-range predictions, and W score is improved by an average of 2 points.

DEMO-EM2: assembling protein complex structures from cryo-EM maps through intertwined chain and domain fitting

The breakthrough in cryo-electron microscopy (cryo-EM) technology has led to an increasing number of density maps of biological macromolecules. However, constructing accurate protein complex atomic structures from cryo-EM maps remains a challenge. In this study, we extend our previously developed DEMO-EM to present DEMO-EM2, an automated method for constructing protein complex models from cryo-EM maps through an iterative assembly procedure intertwining chain- and domain-level matching and fitting for predicted chain models. The method was carefully evaluated on 27 cryo-electron tomography (cryo-ET) maps and 16 single-particle EM maps, where DEMO-EM2 models achieved an average TM-score of 0.92, outperforming those of state-of-the-art methods. The results demonstrate an efficient method that enables the rapid and reliable solution of challenging cryo-EM structure modeling problems.

Integrating deep learning, threading alignments, and a multi-MSA strategy for high-quality protein monomer and complex structure prediction in CASP15

We report the results of the "UM-TBM" and "Zheng" groups in CASP15 for protein monomer and complex structure prediction. These prediction sets were obtained using the D-I-TASSER and DMFold-Multimer algorithms, respectively. For monomer structure prediction, D-I-TASSER introduced four new features during CASP15: (i) a multiple sequence alignment (MSA) generation protocol that combines multi-source MSA searching and a structural modeling-based MSA ranker; (ii) attention-network based spatial restraints; (iii) a multi-domain module containing domain partition and arrangement for domain-level templates and spatial restraints; (iv) an optimized I-TASSER-based folding simulation system for full-length model creation guided by a combination of deep learning restraints, threading alignments, and knowledge-based potentials. For 47 free modeling targets in CASP15, the final models predicted by D-I-TASSER showed average TM-score 19% higher than the standard AlphaFold2 program. We thus showed that traditional Monte Carlo-based folding simulations, when appropriately coupled with deep learning algorithms, can generate models with improved accuracy over end-to-end deep learning methods alone. For protein complex structure prediction, DMFold-Multimer generated models by integrating a new MSA generation algorithm (DeepMSA2) with the end-to-end modeling module from AlphaFold2-Multimer. For the 38 complex targets, DMFold-Multimer generated models with an average TM-score of 0.83 and Interface Contact Score of 0.60, both significantly higher than those of competing complex prediction tools. Our analyses on complexes highlighted the critical role played by MSA generating, ranking, and pairing in protein complex structure prediction. We also discuss future room for improvement in the areas of viral protein modeling and complex model ranking.

Integrating end-to-end learning with deep geometrical potentials for ab initio RNA structure prediction

RNAs are fundamental in living cells and perform critical functions determined by their tertiary architectures. However, accurate modeling of 3D RNA structure remains a challenging problem. We present a novel method, DRfold, to predict RNA tertiary structures by simultaneous learning of local frame rotations and geometric restraints from experimentally solved RNA structures, where the learned knowledge is converted into a hybrid energy potential to guide RNA structure assembly. The method significantly outperforms previous approaches by >73.3% in TM-score on a sequence-nonredundant dataset containing recently released structures. Detailed analyses showed that the major contribution to the improvements arise from the deep end-to-end learning supervised with the atom coordinates and the composite energy function integrating complementary information from geometry restraints and end-to-end learning models. The open-source DRfold program with fast training protocol allows large-scale application of high-resolution RNA structure modeling and can be further improved with future expansion of RNA structure databases.

PIQLE: protein-protein interface quality estimation by deep graph learning of multimeric interaction geometries

Motivation

Accurate modeling of protein-protein interaction interface is essential for high-quality protein complex structure prediction. Existing approaches for estimating the quality of a predicted protein complex structural model utilize only the physicochemical properties or energetic contributions of the interacting atoms, ignoring evolutionarily information or inter-atomic multimeric geometries, including interaction distance and orientations.

Results

Here, we present PIQLE, a deep graph learning method for protein-protein interface quality estimation. PIQLE leverages multimeric interaction geometries and evolutionarily information along with sequence- and structure-derived features to estimate the quality of individual interactions between the interfacial residues using a multi-head graph attention network and then probabilistically combines the estimated quality for scoring the overall interface. Experimental results show that PIQLE consistently outperforms existing state-of-the-art methods including DProQA, TRScore, GNN-DOVE and DOVE on multiple independent test datasets across a wide range of evaluation metrics. Our ablation study and comparison with the self-assessment module of AlphaFold-Multimer repurposed for protein complex scoring reveal that the performance gains are connected to the effectiveness of the multi-head graph attention network in leveraging multimeric interaction geometries and evolutionary information along with other sequence- and structure-derived features adopted in PIQLE.

Availability and implementation

An open-source software implementation of PIQLE is freely available at https://github.com/Bhattacharya-Lab/PIQLE.

Supplementary information

Supplementary data are available at Bioinformatics Advances online.

Coevolutionary signals in metabotropic glutamate receptors capture residue contacts and long-range functional interactions

Upon ligand binding to a G protein-coupled receptor, extracellular signals are transmitted into a cell through sets of residue interactions that translate ligand binding into structural rearrangements. These interactions needed for functions impose evolutionary constraints so that, on occasion, mutations in one position may be compensated by other mutations at functionally coupled positions. To quantify the impact of amino acid substitutions in the context of major evolutionary divergence in the G protein-coupled receptor subfamily of metabotropic glutamate receptors (mGluRs), we combined two phylogenetic-based algorithms, Evolutionary Trace and covariation Evolutionary Trace, to infer potential structure-function couplings and roles in mGluRs. We found a subset of evolutionarily important residues at known functional sites and evidence of coupling among distinct structural clusters in mGluR. In addition, experimental mutagenesis and functional assays confirmed that some highly covariant residues are coupled, revealing their synergy. Collectively, these findings inform a critical step toward understanding the molecular and structural basis of amino acid variation patterns within mGluRs and provide insight for drug development, protein engineering, and analysis of naturally occurring variants.

PIQLE: protein-protein interface quality estimation by deep graph learning of multimeric interaction geometries

Accurate modeling of protein-protein interaction interface is essential for high-quality protein complex structure prediction. Existing approaches for estimating the quality of a predicted protein complex structural model utilize only the physicochemical properties or energetic contributions of the interacting atoms, ignoring evolutionarily information or inter-atomic multimeric geometries, including interaction distance and orientations. Here we present PIQLE, a deep graph learning method for protein-protein interface quality estimation. PIQLE leverages multimeric interaction geometries and evolutionarily information along with sequence- and structure-derived features to estimate the quality of the individual interactions between the interfacial residues using a multihead graph attention network and then probabilistically combines the estimated quality of the interfacial residues for scoring the overall interface. Experimental results show that PIQLE consistently outperforms existing state-of-the-art methods on multiple independent test datasets across a wide range of evaluation metrics. Our ablation study reveals that the performance gains are connected to the effectiveness of the multihead graph attention network in leveraging multimeric interaction geometries and evolutionary information along with other sequence- and structure-derived features adopted in PIQLE. An open-source software implementation of PIQLE, licensed under the GNU General Public License v3, is freely available at https://github.com/Bhattacharya-Lab/PIQLE .

I-DNAN6mA: Accurate Identification of DNA N6-Methyladenine Sites Using the Base-Pairing Map and Deep Learning

The recent discovery of numerous DNA N⁶-methyladenine (6mA) sites has transformed our perception about the roles of 6mA in living organisms. However, our ability to understand them is hampered by our inability to identify 6mA sites rapidly and cost-efficiently by existing experimental methods. Developing a novel method to quickly and accurately identify 6mA sites is critical for speeding up the progress of its function detection and understanding. In this study, we propose a novel computational method, called I-DNAN6mA, to identify 6mA sites and complement experimental methods well, by leveraging the base-pairing rules and a well-designed three-stage deep learning model with pairwise inputs. The performance of our proposed method is benchmarked and evaluated on four species, i.e., Arabidopsis thaliana, Drosophila melanogaster, Rice, and Rosaceae. The experimental results demonstrate that I-DNAN6mA achieves area under the receiver operating characteristic curve values of 0.967, 0.963, 0.947, 0.976, and 0.990, accuracies of 91.5, 92.7, 88.2, 0.938, and 96.2%, and Mathew's correlation coefficient values of 0.855, 0.831, 0.763, 0.877, and 0.924 on five benchmark data sets, respectively, and outperforms several existing state-of-the-art methods. To our knowledge, I-DNAN6mA is the first approach to identify 6mA sites using a novel image-like representation of DNA sequences and a deep learning model with pairwise inputs. I-DNAN6mA is expected to be useful for locating functional regions of DNA.

Prediction of the tetramer protein complex interaction based on CNN and SVM

Protein-protein interactions play an important role in life activities. The study of protein-protein interactions helps to better understand the mechanism of protein complex interaction, which is crucial for drug design, protein function annotation and three-dimensional structure prediction of protein complexes. In this paper, we study the tetramer protein complex interaction. The research has two parts: The first part is to predict the interaction between chains of the tetramer protein complex. In this part, we proposed a feature map to represent a sample generated by two chains of the tetramer protein complex, and constructed a Convolutional Neural Network (CNN) model to predict the interaction between chains of the tetramer protein complex. The AUC value of testing set is 0.6263, which indicates that our model can be used to predict the interaction between chains of the tetramer protein complex. The second part is to predict the tetramer protein complex interface residue pairs. In this part, we proposed a Support Vector Machine (SVM) ensemble method based on under-sampling and ensemble method to predict the tetramer protein complex interface residue pairs. In the top 10 predictions, when at least one protein-protein interaction interface is correctly predicted, the accuracy of our method is 82.14%. The result shows that our method is effective for the prediction of the tetramer protein complex interface residue pairs.

DeepHomo2.0: improved protein-protein contact prediction of homodimers by transformer-enhanced deep learning

Protein-protein interactions play an important role in many biological processes. However, although structure prediction for monomer proteins has achieved great progress with the advent of advanced deep learning algorithms like AlphaFold, the structure prediction for protein-protein complexes remains an open question. Taking advantage of the Transformer model of ESM-MSA, we have developed a deep learning-based model, named DeepHomo2.0, to predict protein-protein interactions of homodimeric complexes by leveraging the direct-coupling analysis (DCA) and Transformer features of sequences and the structure features of monomers. DeepHomo2.0 was extensively evaluated on diverse test sets and compared with eight state-of-the-art methods including protein language model-based, DCA-based and machine learning-based methods. It was shown that DeepHomo2.0 achieved a high precision of >70% with experimental monomer structures and >60% with predicted monomer structures for the top 10 predicted contacts on the test sets and outperformed the other eight methods. Moreover, even the version without using structure information, named DeepHomoSeq, still achieved a good precision of >55% for the top 10 predicted contacts. Integrating the predicted contacts into protein docking significantly improved the structure prediction of realistic Critical Assessment of Protein Structure Prediction homodimeric complexes. DeepHomo2.0 and DeepHomoSeq are available at http://huanglab.phys.hust.edu.cn/DeepHomo2/.

Contact-Assisted Threading in Low-Homology Protein Modeling

The ability to successfully predict the three-dimensional structure of a protein from its amino acid sequence has made considerable progress in the recent past. The progress is propelled by the improved accuracy of deep learning-based inter-residue contact map predictors coupled with the rising growth of protein sequence databases. Contact map encodes interatomic interaction information that can be exploited for highly accurate prediction of protein structures via contact map threading even for the query proteins that are not amenable to direct homology modeling. As such, contact-assisted threading has garnered considerable research effort. In this chapter, we provide an overview of existing contact-assisted threading methods while highlighting the recent advances and discussing some of the current limitations and future prospects in the application of contact-assisted threading for improving the accuracy of low-homology protein modeling.

rrQNet: Protein contact map quality estimation by deep evolutionary reconciliation

Protein contact maps have proven to be a valuable tool in the deep learning revolution of protein structure prediction, ushering in the recent breakthrough by AlphaFold2. However, self-assessment of the quality of predicted structures are typically performed at the granularity of three-dimensional coordinates as opposed to directly exploiting the rotation- and translation-invariant two-dimensional (2D) contact maps. Here, we present rrQNet, a deep learning method for self-assessment in 2D by contact map quality estimation. Our approach is based on the intuition that for a contact map to be of high quality, the residue pairs predicted to be in contact should be mutually consistent with the evolutionary context of the protein. The deep neural network architecture of rrQNet implements this intuition by cascading two deep modules-one encoding the evolutionary context and the other performing evolutionary reconciliation. The penultimate stage of rrQNet estimates the quality scores at the interacting residue-pair level, which are then aggregated for estimating the quality of a contact map. This design choice offers versatility at varied resolutions from individual residue pairs to full-fledged contact maps. Trained on multiple complementary sources of contact predictors, rrQNet facilitates generalizability across various contact maps. By rigorously testing using publicly available datasets and comparing against several in-house baseline approaches, we show that rrQNet accurately reproduces the true quality score of a predicted contact map and successfully distinguishes between accurate and inaccurate contact maps predicted by a wide variety of contact predictors. The open-source rrQNet software package is freely available at https://github.com/Bhattacharya-Lab/rrQNet.

Prediction of inter-chain distance maps of protein complexes with 2D attention-based deep neural networks

Residue-residue distance information is useful for predicting tertiary structures of protein monomers or quaternary structures of protein complexes. Many deep learning methods have been developed to predict intra-chain residue-residue distances of monomers accurately, but few methods can accurately predict inter-chain residue-residue distances of complexes. We develop a deep learning method CDPred (i.e., Complex Distance Prediction) based on the 2D attention-powered residual network to address the gap. Tested on two homodimer datasets, CDPred achieves the precision of 60.94% and 42.93% for top L/5 inter-chain contact predictions (L: length of the monomer in homodimer), respectively, substantially higher than DeepHomo's 37.40% and 23.08% and GLINTER's 48.09% and 36.74%. Tested on the two heterodimer datasets, the top Ls/5 inter-chain contact prediction precision (Ls: length of the shorter monomer in heterodimer) of CDPred is 47.59% and 22.87% respectively, surpassing GLINTER's 23.24% and 13.49%. Moreover, the prediction of CDPred is complementary with that of AlphaFold2-multimer.

Res-Dom: predicting protein domain boundary from sequence using deep residual network and Bi-LSTM

Motivation

Protein domains are the basic units of proteins that can fold, function and evolve independently. Protein domain boundary partition plays an important role in protein structure prediction, understanding their biological functions, annotating their evolutionary mechanisms and protein design. Although there are many methods that have been developed to predict domain boundaries from protein sequence over the past two decades, there is still much room for improvement.

Results

In this article, a novel domain boundary prediction tool called Res-Dom was developed, which is based on a deep residual network, bidirectional long short-term memory (Bi-LSTM) and transfer learning. We used deep residual neural networks to extract higher-order residue-related information. In addition, we also used a pre-trained protein language model called ESM to extract sequence embedded features, which can summarize sequence context information more abundantly. To improve the global representation of these deep residual networks, a Bi-LSTM network was also designed to consider long-range interactions between residues. Res-Dom was then tested on an independent test set including 342 proteins and generated correct single-domain and multi-domain classifications with a Matthew's correlation coefficient of 0.668, which was 17.6% higher than the second-best compared method. For domain boundaries, the normalized domain overlapping score of Res-Dom was 0.849, which was 5% higher than the second-best compared method. Furthermore, Res-Dom required significantly less time than most of the recently developed state-of-the-art domain prediction methods.

Availability and implementation

All source code, datasets and model are available at http://isyslab.info/Res-Dom/.

ResSUMO: A Deep Learning Architecture Based on Residual Structure for Prediction of Lysine SUMOylation Sites

Lysine SUMOylation plays an essential role in various biological functions. Several approaches integrating various algorithms have been developed for predicting SUMOylation sites based on a limited dataset. Recently, the number of identified SUMOylation sites has significantly increased due to investigation at the proteomics scale. We collected modification data and found the reported approaches had poor performance using our collected data. Therefore, it is essential to explore the characteristics of this modification and construct prediction models with improved performance based on an enlarged dataset. In this study, we constructed and compared 16 classifiers by integrating four different algorithms and four encoding features selected from 11 sequence-based or physicochemical features. We found that the convolution neural network (CNN) model integrated with residue structure, dubbed ResSUMO, performed favorably when compared with the traditional machine learning and CNN models in both cross-validation and independent tests. The area under the receiver operating characteristic (ROC) curve for ResSUMO was around 0.80, superior to that of the reported predictors. We also found that increasing the depth of neural networks in the CNN models did not improve prediction performance due to the degradation problem, but the residual structure could be included to optimize the neural networks and improve performance. This indicates that residual neural networks have the potential to be broadly applied in the prediction of other types of modification sites with great effectiveness and robustness. Furthermore, the online ResSUMO service is freely accessible.

I-TASSER-MTD: a deep-learning-based platform for multi-domain protein structure and function prediction

Most proteins in cells are composed of multiple folding units (or domains) to perform complex functions in a cooperative manner. Relative to the rapid progress in single-domain structure prediction, there are few effective tools available for multi-domain protein structure assembly, mainly due to the complexity of modeling multi-domain proteins, which involves higher degrees of freedom in domain-orientation space and various levels of continuous and discontinuous domain assembly and linker refinement. To meet the challenge and the high demand of the community, we developed I-TASSER-MTD to model the structures and functions of multi-domain proteins through a progressive protocol that combines sequence-based domain parsing, single-domain structure folding, inter-domain structure assembly and structure-based function annotation in a fully automated pipeline. Advanced deep-learning models have been incorporated into each of the steps to enhance both the domain modeling and inter-domain assembly accuracy. The protocol allows for the incorporation of experimental cross-linking data and cryo-electron microscopy density maps to guide the multi-domain structure assembly simulations. I-TASSER-MTD is built on I-TASSER but substantially extends its ability and accuracy in modeling large multi-domain protein structures and provides meaningful functional insights for the targets at both the domain- and full-chain levels from the amino acid sequence alone.

cpxDeepMSA: A Deep Cascade Algorithm for Constructing Multiple Sequence Alignments of Protein–Protein Interactions

Protein–protein interactions (PPIs) are fundamental to many biological processes. The coevolution-based prediction of interacting residues has made great strides in protein complexes that are known to interact. A multiple sequence alignment (MSA) is the basis of coevolution analysis. MSAs have recently made significant progress in the protein monomer sequence analysis. However, no standard or efficient pipelines are available for the sensitive protein complex MSA (cpxMSA) collection. How to generate cpxMSA is one of the most challenging problems of sequence coevolution analysis. Although several methods have been developed to address this problem, no standalone program exists. Furthermore, the number of built-in properties is limited; hence, it is often difficult for users to analyze sequence coevolution according to their desired cpxMSA. In this article, we developed a novel cpxMSA approach (cpxDeepMSA. We used different protein monomer databases and incorporated the three strategies (genomic distance, phylogeny information, and STRING interaction network) used to join the monomer MSA results of protein complexes, which can prevent using a single method fail to the joint two-monomer MSA causing the cpxMSA construction failure. We anticipate that the cpxDeepMSA algorithm will become a useful high-throughput tool in protein complex structure predictions, inter-protein residue-residue contacts, and the biological sequence coevolution analysis.

SPROUT: spectral sparsification helps restore the spatial structure at single-cell resolution

Single-cell RNA sequencing thoroughly quantifies the individual cell transcriptomes but renounces the spatial structure. Conversely, recently emerged spatial transcriptomics technologies capture the cellular spatial structure but skimp cell or gene resolutions. Ligand-receptor interactions reveal the potential of cell proximity since they are spatially constrained. Cell–cell affinity values estimated by ligand–receptor interaction can partially represent the structure of cells but falsely include the pseudo affinities between distant or indirectly interacting cells. Here, we develop a software package, SPROUT, to reconstruct the single-cell resolution spatial structure from the transcriptomics data through diminished pseudo ligand–receptor affinities. For spatial data, SPROUT first curates the representative single-cell profiles for each spatial spot from a candidate library, then reduces the pseudo affinities in the intercellular affinity matrix by partial correlation, spectral graph sparsification, and spatial coordinates refinement. SPROUT embeds the estimated interactions into a low-dimensional space with the cross-entropy objective to restore the intercellular structures, which facilitates the discovery of dominant ligand–receptor pairs between neighboring cells at single-cell resolution. SPROUT reconstructed structures achieved shape Pearson correlations ranging from 0.91 to 0.97 on the mouse hippocampus and human organ tumor microenvironment datasets. Furthermore, SPROUT can solely de novo reconstruct the structures at single-cell resolution, i.e., reaching the cell-type proximity correlations of 0.68 and 0.89 between reconstructed and immunohistochemistry-informed spatial structures on a human developing heart dataset and a tumor microenvironment dataset, respectively.

Deep learning geometrical potential for high-accuracy ab initio protein structure prediction

Ab initio protein structure prediction has been vastly boosted by the modeling of inter-residue contact/distance maps in recent years. We developed a new deep learning model, DeepPotential, which accurately predicts the distribution of a complementary set of geometric descriptors including a novel hydrogen-bonding potential defined by C-alpha atom coordinates. On 154 Free-Modeling/Hard targets from the CASP and CAMEO experiments, DeepPotential demonstrated significant advantage on both geometrical feature prediction and full-length structure construction, with Top-L/5 contact accuracy and TM-score of full-length models 4.1% and 6.7% higher than the best of other deep-learning restraint prediction approaches. Detail analyses showed that the major contributions to the TM-score/contact-map improvements come from the employment of multi-tasking network architecture and metagenome-based MSA collection assisted with confidence-based MSA selection, where hydrogen-bonding and inter-residue orientation predictions help improve hydrogen-bonding network and secondary structure packing. These results demonstrated new progress in the deep-learning restraint-guided ab initio protein structure prediction.

•

Multi-tasking network architecture for multiple inter-residue geometries

•

Novel deep learning model for improved hydrogen-bonding modeling

•

Rapid and high-accuracy Ab initio protein structure prediction

BIAPSS: A Comprehensive Physicochemical Analyzer of Proteins Undergoing Liquid-Liquid Phase Separation

The liquid-liquid phase separation (LLPS) of biomolecules is a phenomenon which is nowadays recognized as the driving force for the biogenesis of numerous functional membraneless organelles and cellular bodies. The interplay between the protein primary sequence and phase separation remains poorly understood, despite intensive research. To uncover the sequence-encoded signals of protein capable of undergoing LLPS, we developed a novel web platform named BIAPSS (Bioinformatics Analysis of LLPS Sequences). This web server provides on-the-fly analysis, visualization, and interpretation of the physicochemical and structural features for the superset of curated LLPS proteins.

Inter-Residue Distance Prediction From Duet Deep Learning Models

Residue distance prediction from the sequence is critical for many biological applications such as protein structure reconstruction, protein–protein interaction prediction, and protein design. However, prediction of fine-grained distances between residues with long sequence separations still remains challenging. In this study, we propose DuetDis, a method based on duet feature sets and deep residual network with squeeze-and-excitation (SE), for protein inter-residue distance prediction. DuetDis embraces the ability to learn and fuse features directly or indirectly extracted from the whole-genome/metagenomic databases and, therefore, minimize the information loss through ensembling models trained on different feature sets. We evaluate DuetDis and 11 widely used peer methods on a large-scale test set (610 proteins chains). The experimental results suggest that 1) prediction results from different feature sets show obvious differences; 2) ensembling different feature sets can improve the prediction performance; 3) high-quality multiple sequence alignment (MSA) used for both training and testing can greatly improve the prediction performance; and 4) DuetDis is more accurate than peer methods for the overall prediction, more reliable in terms of model prediction score, and more robust against shallow multiple sequence alignment (MSA).

Mimetic Neural Networks: A Unified Framework for Protein Design and Folding

Recent advancements in machine learning techniques for protein structure prediction motivate better results in its inverse problem–protein design. In this work we introduce a new graph mimetic neural network, MimNet, and show that it is possible to build a reversible architecture that solves the structure and design problems in tandem, allowing to improve protein backbone design when the structure is better estimated. We use the ProteinNet data set and show that the state of the art results in protein design can be met and even improved, given recent architectures for protein folding.

Performing protein fold recognition by exploiting a stack convolutional neural network with the attention mechanism

Protein fold recognition is a critical step in protein structure and function prediction, and aims to ascertain the most likely fold type of the query protein. As a typical pattern recognition problem, designing a powerful feature extractor and metric function to extract relevant and representative fold-specific features from protein sequences is the key to improving protein fold recognition. In this study, we propose an effective sequence-based approach, called RattnetFold, to identify protein fold types. The basic concept of RattnetFold is to employ a stack convolutional neural network with the attention mechanism that acts as a feature extractor to extract fold-specific features from protein residue-residue contact maps. Moreover, based on the fold-specific features, we leverage metric learning to project fold-specific features into a subspace where similar proteins are closer together and name this approach RattnetFoldPro. Benchmarking experiments illustrate that RattnetFold and RattnetFoldPro enable the convolutional neural networks to efficiently learn the underlying subtle patterns in residue-residue contact maps, thereby improving the performance of protein fold recognition. An online web server of RattnetFold and the benchmark datasets are freely available at http://csbio.njust.edu.cn/bioinf/rattnetfold/.

Progressive assembly of multi-domain protein structures from cryo-EM density maps

Progress in cryo-electron microscopy has provided the potential for large-size protein structure determination. However, the success rate for solving multi-domain proteins remains low because of the difficulty in modelling inter-domain orientations. Here we developed domain enhanced modeling using cryo-electron microscopy (DEMO-EM), an automatic method to assemble multi-domain structures from cryo-electron microscopy maps through a progressive structural refinement procedure combining rigid-body domain fitting and flexible assembly simulations with deep-neural-network inter-domain distance profiles. The method was tested on a large-scale benchmark set of proteins containing up to 12 continuous and discontinuous domains with medium- to low-resolution density maps, where DEMO-EM produced models with correct inter-domain orientations (template modeling score (TM-score) >0.5) for 97% of cases and outperformed state-of-the-art methods. DEMO-EM was applied to the severe acute respiratory syndrome coronavirus 2 genome and generated models with average TM-score and root-mean-square deviation of 0.97 and 1.3 Å, respectively, with respect to the deposited structures. These results demonstrate an efficient pipeline that enables automated and reliable large-scale multi-domain protein structure modelling from cryo-electron microscopy maps.

Deep learning methods for 3D structural proteome and interactome modeling

Bolstered by recent methodological and hardware advances, deep learning has increasingly been applied to biological problems and structural proteomics. Such approaches have achieved remarkable improvements over traditional machine learning methods in tasks ranging from protein contact map prediction to protein folding, prediction of protein-protein interaction interfaces, and characterization of protein-drug binding pockets. In particular, emergence of ab initio protein structure prediction methods including AlphaFold2 has revolutionized protein structural modeling. From a protein function perspective, numerous deep learning methods have facilitated deconvolution of the exact amino acid residues and protein surface regions responsible for binding other proteins or small molecule drugs. In this review, we provide a comprehensive overview of recent deep learning methods applied in structural proteomics.

The Structural Determinants of Intra-Protein Compensatory Substitutions

Compensatory substitutions happen when one mutation is advantageously selected because it restores the loss of fitness induced by a previous deleterious mutation. How frequent such mutations occur in evolution and what is the structural and functional context permitting their emergence remain open questions. We built an atlas of intra-protein compensatory substitutions using a phylogenetic approach and a dataset of 1,630 bacterial protein families for which high-quality sequence alignments and experimentally derived protein structures were available. We identified more than 51,000 positions coevolving by the mean of predicted compensatory mutations. Using the evolutionary and structural properties of the analyzed positions, we demonstrate that compensatory mutations are scarce (typically only a few in the protein history) but widespread (the majority of proteins experienced at least one). Typical coevolving residues are evolving slowly, are located in the protein core outside secondary structure motifs, and are more often in contact than expected by chance, even after accounting for their evolutionary rate and solvent exposure. An exception to this general scheme is residues coevolving for charge compensation, which are evolving faster than noncoevolving sites, in contradiction with predictions from simple coevolutionary models, but similar to stem pairs in RNA. While sites with a significant pattern of coevolution by compensatory mutations are rare, the comparative analysis of hundreds of structures ultimately permits a better understanding of the link between the three-dimensional structure of a protein and its fitness landscape.

Protein design via deep learning

Proteins with desired functions and properties are important in fields like nanotechnology and biomedicine. De novo protein design enables the production of previously unseen proteins from the ground up and is believed as a key point for handling real social challenges. Recent introduction of deep learning into design methods exhibits a transformative influence and is expected to represent a promising and exciting future direction. In this review, we retrospect the major aspects of current advances in deep-learning-based design procedures and illustrate their novelty in comparison with conventional knowledge-based approaches through noticeable cases. We not only describe deep learning developments in structure-based protein design and direct sequence design, but also highlight recent applications of deep reinforcement learning in protein design. The future perspectives on design goals, challenges and opportunities are also comprehensively discussed.

CR-I-TASSER: assemble protein structures from cryo-EM density maps using deep convolutional neural networks

Cryo-electron microscopy (cryo-EM) has become a leading approach for protein structure determination, but it remains challenging to accurately model atomic structures with cryo-EM density maps. We propose a hybrid method, CR-I-TASSER, which integrates deep neural-network learning with I-TASSER assembly simulations for automated cryo-EM structure determination. The method is benchmarked on 778 proteins with simulated and experimental density maps, where CR-I-TASSER constructs models with a correct fold (TM-score>0.5) for 643 targets that is 64% higher than the best of other de novo and refinement-based approaches on high-resolution data samples. Detailed data analyses showed that the major advantage of CR-I-TASSER lies in the deep-learning based Cα position prediction, which significantly improves the threading template quality and therefore boosts the accuracy of final models through optimized fragment assembly simulations. These results demonstrate a new avenue to determine cryo-EM protein structures with high accuracy and robustness covering various target types and density-map resolutions.

DisCovER: distance- and orientation-based covariational threading for weakly homologous proteins

Threading a query protein sequence onto a library of weakly homologous structural templates remains challenging, even when sequence-based predicted contact or distance information is used. Contact-assisted or distance-assisted threading methods utilize only the spatial proximity of the interacting residue pairs for template selection and alignment, ignoring their orientation. Moreover, existing threading methods fail to consider the neighborhood effect induced by the query-template alignment. We present a new distance- and orientation-based covariational threading method called DisCovER by effectively integrating information from inter-residue distance and orientation along with the topological network neighborhood of a query-template alignment. Our method first selects a subset of templates using standard profile-based threading coupled with topological network similarity terms to account for the neighborhood effect and subsequently performs distance- and orientation-based query-template alignment using an iterative double dynamic programming framework. Multiple large-scale benchmarking results on query proteins classified as weakly homologous from the continuous automated model evaluation experiment and from the current literature show that our method outperforms several existing state-of-the-art threading approaches, and that the integration of the neighborhood effect with the inter-residue distance and orientation information synergistically contributes to the improved performance of DisCovER. DisCovER is freely available at https://github.com/Bhattacharya-Lab/DisCovER.

How sticky are our proteins? Quantifying hydrophobicity of the human proteome

Summary

Proteins tend to bury hydrophobic residues inside their core during the folding process to provide stability to the protein structure and to prevent aggregation. Nevertheless, proteins do expose some 'sticky' hydrophobic residues to the solvent. These residues can play an important functional role, e.g. in protein-protein and membrane interactions. Here, we first investigate how hydrophobic protein surfaces are by providing three measures for surface hydrophobicity: the total hydrophobic surface area, the relative hydrophobic surface area and-using our MolPatch method-the largest hydrophobic patch. Secondly, we analyze how difficult it is to predict these measures from sequence: by adapting solvent accessibility predictions from NetSurfP2.0, we obtain well-performing prediction methods for the THSA and RHSA, while predicting LHP is more challenging. Finally, we analyze implications of exposed hydrophobic surfaces: we show that hydrophobic proteins typically have low expression, suggesting cells avoid an overabundance of sticky proteins.

Availability and implementation

The data underlying this article are available in GitHub at https://github.com/ibivu/hydrophobic_patches.

Supplementary information

Supplementary data are available at Bioinformatics Advances online.

Enhancing protein inter-residue real distance prediction by scrutinising deep learning models

Protein structure prediction (PSP) has achieved significant progress lately via prediction of inter-residue distances using deep learning models and exploitation of the predictions during conformational search. In this context, prediction of large inter-residue distances and also prediction of distances between residues separated largely in the protein sequence remain challenging. To deal with these challenges, state-of-the-art inter-residue distance prediction algorithms have used large sets of coevolutionary and non-coevolutionary features. In this paper, we argue that the more the types of features used, the more the kinds of noises introduced and then the deep learning model has to overcome the noises to improve the accuracy of the predictions. Also, multiple features capturing similar underlying characteristics might not necessarily have significantly better cumulative effect. So we scrutinise the feature space to reduce the types of features to be used, but at the same time, we strive to improve the prediction accuracy. Consequently, for inter-residue real distance prediction, in this paper, we propose a deep learning model named scrutinised distance predictor (SDP), which uses only 2 coevolutionary and 3 non-coevolutionary features. On several sets of benchmark proteins, our proposed SDP method improves mean Local Distance Different Test (LDDT) scores at least by 10% over existing state-of-the-art methods. The SDP program along with its data is available from the website https://gitlab.com/mahnewton/sdp .

OUP accepted manuscript

Motivation

Deep learning has revolutionized protein tertiary structure prediction recently. The cutting-edge deep learning methods such as AlphaFold can predict high-accuracy tertiary structures for most individual protein chains. However, the accuracy of predicting quaternary structures of protein complexes consisting of multiple chains is still relatively low due to lack of advanced deep learning methods in the field. Because interchain residue–residue contacts can be used as distance restraints to guide quaternary structure modeling, here we develop a deep dilated convolutional residual network method (DRCon) to predict interchain residue–residue contacts in homodimers from residue–residue co-evolutionary signals derived from multiple sequence alignments of monomers, intrachain residue–residue contacts of monomers extracted from true/predicted tertiary structures or predicted by deep learning, and other sequence and structural features.

Results

Tested on three homodimer test datasets (Homo_std dataset, DeepHomo dataset and CASP-CAPRI dataset), the precision of DRCon for top L/5 interchain contact predictions (L: length of monomer in a homodimer) is 43.46%, 47.10% and 33.50% respectively at 6 Å contact threshold, which is substantially better than DeepHomo and DNCON2_inter and similar to Glinter. Moreover, our experiments demonstrate that using predicted tertiary structure or intrachain contacts of monomers in the unbound state as input, DRCon still performs well, even though its accuracy is lower than using true tertiary structures in the bound state are used as input. Finally, our case study shows that good interchain contact predictions can be used to build high-accuracy quaternary structure models of homodimers.

Availability and implementation

The source code of DRCon is available at https://github.com/jianlin-cheng/DRCon. The datasets are available at https://zenodo.org/record/5998532#.YgF70vXMKsB.

Supplementary information

Supplementary data are available at Bioinformatics online.

De novo Protein Structure Prediction by Coupling Contact With Distance Profile

De novo protein structure prediction is a challenging problem that requires both an accurate energy function and an efficient conformation sampling method. In this study, a de novo structure prediction method, named CoDiFold, is proposed. In CoDiFold, contacts and distance profiles are organically combined into the Rosetta low-resolution energy function to improve the accuracy of energy function. As a result, the correlation between energy and root mean square deviation (RMSD) is improved. In addition, a population-based multi-mutation strategy is designed to balance the exploration and exploitation of conformation space sampling. The average RMSD of the models generated by the proposed protocol is decreased by 49.24 and 45.21 percent in the test set with 43 proteins compared with those of Rosetta and QUARK de novo protocols, respectively. The results also demonstrate that the structures predicted by proposed CoDiFold are comparable to the state-of-the-art methods for the 10 FM targets of CASP13. The source code and executable versions are freely available at http://github.com/iobio-zjut/CoDiFold.

A de novo protein structure prediction by iterative partition sampling, topology adjustment and residue-level distance deviation optimization

MOTIVATION

With the great progress of deep learning-based inter-residue contact/distance prediction, the discrete space formed by fragment assembly cannot satisfy the distance constraint well. Thus, the optimal solution of the continuous space may not be achieved. Designing an effective closed-loop continuous dihedral angle optimization strategy that complements the discrete fragment assembly is crucial to improve the performance of the distance-assisted fragment assembly method.

RESULTS

In this article, we proposed a de novo protein structure prediction method called IPTDFold based on closed-loop iterative partition sampling, topology adjustment and residue-level distance deviation optimization. First, local dihedral angle crossover and mutation operators are designed to explore the conformational space extensively and achieve information exchange between the conformations in the population. Then, the dihedral angle rotation model of loop region with partial inter-residue distance constraints is constructed, and the rotation angle satisfying the constraints is obtained by differential evolution algorithm, so as to adjust the spatial position relationship between the secondary structures. Finally, the residue distance deviation is evaluated according to the difference between the conformation and the predicted distance, and the dihedral angle of the residue is optimized with biased probability. The final model is generated by iterating the above three steps. IPTDFold is tested on 462 benchmark proteins, 24 FM targets of CASP13 and 20 FM targets of CASP14. Results show that IPTDFold is significantly superior to the distance-assisted fragment assembly method Rosetta_D (Rosetta with distance). In particular, the prediction accuracy of IPTDFold does not decrease as the length of the protein increases. When using the same FastRelax protocol, the prediction accuracy of IPTDFold is significantly superior to that of trRosetta without orientation constraints, and is equivalent to that of the full version of trRosetta.

AVAILABILITYAND IMPLEMENTATION

The source code and executable are freely available at https://github.com/iobio-zjut/IPTDFold.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Decoding the link of microbiome niches with homologous sequences enables accurately targeted protein structure prediction

Information derived from metagenome sequences through deep-learning techniques has significantly improved the accuracy of template free protein structure modeling. However, most of the deep learning-based modeling studies are based on blind sequence database searches and suffer from low efficiency in computational resource utilization and model construction, especially when the sequence library becomes prohibitively large. We proposed a MetaSource model built on 4.25 billion microbiome sequences from four major biomes (Gut, Lake, Soil, and Fermentor) to decode the inherent linkage of microbial niches with protein homologous families. Large-scale protein family folding experiments on 8,700 unknown Pfam families showed that a microbiome targeted approach with multiple sequence alignment constructed from individual MetaSource biomes requires more than threefold less computer memory and CPU (central processing unit) time but generates contact-map and three-dimensional structure models with a significantly higher accuracy, compared with that using combined metagenome datasets. These results demonstrate an avenue to bridge the gap between the rapidly increasing metagenome databases and the limited computing resources for efficient genome-wide database mining, which provides a useful bluebook to guide future microbiome sequence database and modeling development for high-accuracy protein structure and function prediction.

Multi contact-based folding method for de novo protein structure prediction

Meta contact, which combines different contact maps into one to improve contact prediction accuracy and effectively reduce the noise from a single contact map, is a widely used method. However, protein structure prediction using meta contact cannot fully exploit the information carried by original contact maps. In this work, a multi contact-based folding method under the evolutionary algorithm framework, MultiCFold, is proposed. In MultiCFold, the thorough information of different contact maps is directly used by populations to guide protein structure folding. In addition, noncontact is considered as an effective supplement to contact information and can further assist protein folding. MultiCFold is tested on a set of 120 nonredundant proteins, and the average TM-score and average RMSD reach 0.617 and 5.815 Å, respectively. Compared with the meta contact-based method, MetaCFold, average TM-score and average RMSD have a 6.62 and 8.82% improvement. In particular, the import of noncontact information increases the average TM-score by 6.30%. Furthermore, MultiCFold is compared with four state-of-the-art methods of CASP13 on the 24 FM targets, and results show that MultiCFold is significantly better than other methods after the full-atom relax procedure.

Improving protein fold recognition using triplet network and ensemble deep learning

Protein fold recognition is a critical step toward protein structure and function prediction, aiming at providing the most likely fold type of the query protein. In recent years, the development of deep learning (DL) technique has led to massive advances in this important field, and accordingly, the sensitivity of protein fold recognition has been dramatically improved. Most DL-based methods take an intermediate bottleneck layer as the feature representation of proteins with new fold types. However, this strategy is indirect, inefficient and conditional on the hypothesis that the bottleneck layer's representation is assumed as a good representation of proteins with new fold types. To address the above problem, in this work, we develop a new computational framework by combining triplet network and ensemble DL. We first train a DL-based model, termed FoldNet, which employs triplet loss to train the deep convolutional network. FoldNet directly optimizes the protein fold embedding itself, making the proteins with the same fold types be closer to each other than those with different fold types in the new protein embedding space. Subsequently, using the trained FoldNet, we implement a new residue-residue contact-assisted predictor, termed FoldTR, which improves protein fold recognition. Furthermore, we propose a new ensemble DL method, termed FSD_XGBoost, which combines protein fold embedding with the other two discriminative fold-specific features extracted by two DL-based methods SSAfold and DeepFR. The Top 1 sensitivity of FSD_XGBoost increases to 74.8% at the fold level, which is ~9% higher than that of the state-of-the-art method. Together, the results suggest that fold-specific features extracted by different DL methods complement with each other, and their combination can further improve fold recognition at the fold level. The implemented web server of FoldTR and benchmark datasets are publicly available at http://csbio.njust.edu.cn/bioinf/foldtr/.

Improved protein contact prediction using dimensional hybrid residual networks and singularity enhanced loss function

Deep residual learning has shown great success in protein contact prediction. In this study, a new deep residual learning-based protein contact prediction model was developed. Comparing with previous models, a new type of residual block hybridizing 1D and 2D convolutions was designed to increase the effective receptive field of the residual network, and a new loss function emphasizing the easily misclassified residue pairs was proposed to enhance the model training. The developed protein contact prediction model referred to as DRN-1D2D was first evaluated on 105 CASP11 targets, 76 CAMEO hard targets and 398 membrane proteins together with two in house-developed reference models based on either the standard 2D residual block or the traditional BCE loss function, from which we confirmed that both the dimensional hybrid residual block and the singularity enhanced loss function can be employed to improve the model performance for protein contact prediction. DRN-1D2D was further evaluated on 39 CASP13 and CASP14 free modeling targets together with the two reference models and six state-of-the-art protein contact prediction models including DeepCov, DeepCon, DeepConPred2, SPOT-Contact, RaptorX-Contact and TripleRes. The result shows that DRN-1D2D consistently achieved the best performance among all these models.

AlphaFold 2: Why It Works and Its Implications for Understanding the Relationships of Protein Sequence, Structure, and Function

AlphaFold 2 (AF2) was the star of CASP14, the last biannual structure prediction experiment. Using novel deep learning, AF2 predicted the structures of many difficult protein targets at or near experimental resolution. Here, we present our perspective of why AF2 works and show that it is a very sophisticated fold recognition algorithm that exploits the completeness of the library of single domain PDB structures. It has also learned local side chain packing rearrangements that enable it to refine proteins to high resolution. The benefits and limitations of its ability to predict the structures of many more proteins at or close to atomic detail are discussed.

FALCON2: a web server for high-quality prediction of protein tertiary structures

BACKGROUND

Accurate prediction of protein tertiary structures is highly desired as the knowledge of protein structures provides invaluable insights into protein functions. We have designed two approaches to protein structure prediction, including a template-based modeling approach (called ProALIGN) and an ab initio prediction approach (called ProFOLD). Briefly speaking, ProALIGN aligns a target protein with templates through exploiting the patterns of context-specific alignment motifs and then builds the final structure with reference to the homologous templates. In contrast, ProFOLD uses an end-to-end neural network to estimate inter-residue distances of target proteins and builds structures that satisfy these distance constraints. These two approaches emphasize different characteristics of target proteins: ProALIGN exploits structure information of homologous templates of target proteins while ProFOLD exploits the co-evolutionary information carried by homologous protein sequences. Recent progress has shown that the combination of template-based modeling and ab initio approaches is promising.

RESULTS

In the study, we present FALCON2, a web server that integrates ProALIGN and ProFOLD to provide high-quality protein structure prediction service. For a target protein, FALCON2 executes ProALIGN and ProFOLD simultaneously to predict possible structures and selects the most likely one as the final prediction result. We evaluated FALCON2 on widely-used benchmarks, including 104 CASP13 (the 13th Critical Assessment of protein Structure Prediction) targets and 91 CASP14 targets. In-depth examination suggests that when high-quality templates are available, ProALIGN is superior to ProFOLD and in other cases, ProFOLD shows better performance. By integrating these two approaches with different emphasis, FALCON2 server outperforms the two individual approaches and also achieves state-of-the-art performance compared with existing approaches.

CONCLUSIONS

By integrating template-based modeling and ab initio approaches, FALCON2 provides an easy-to-use and high-quality protein structure prediction service for the community and we expect it to enable insights into a deep understanding of protein functions.

Improved 3-D Protein Structure Predictions using Deep ResNet Model

Protein Structure Prediction (PSP) is considered to be a complicated problem in computational biology. In spite of, the remarkable progress made by the co-evolution-based method in PSP, it is still a challenging and unresolved problem. Recently, along with co-evolutionary relationships, deep learning approaches have been introduced in PSP that lead to significant progress. In this paper a novel methodology using deep ResNet architecture for predicting inter-residue distance and dihedral angles is proposed, that aims to generate 125 homologous sequences in an average from a set of customized sequence database. These sequences are used to generate input features. As an outcome of neural networks, a pool of structures is generated from which the lowest potential structure is chosen as the final predicted 3-D protein structure. The proposed method is trained using 6521 protein sequences extracted from Protein Data Bank (PDB). For testing 48 protein sequences whose residue length is less than 400 residues are chosen from the 13th Critical Assessment of protein Structure Prediction (CASP 13) dataset are used. The model is compared with Alphafold, Zhang, and RaptorX. The template modeling (TM) score is used to evaluate the accuracy of the estimated structure. The proposed method produces better performances for 52% of the target sequences while that of Alphafold, Zhang, RaptorX were 10%, 22.9%, and 6% respectively. Additionally, for 37.5% target sequences, the proposed method was able to achieve accuracy greater than or equal to 0.80. The TM score obtained for the sequences under consideration were 0.69, 0.67, 0.65, and 0.58 respectively for the proposed method, Alphafold, Zhang, and RaptorX.

Protein sequence-to-structure learning: Is this the end(-to-end revolution)?

The potential of deep learning has been recognized in the protein structure prediction community for some time, and became indisputable after CASP13. In CASP14, deep learning has boosted the field to unanticipated levels reaching near-experimental accuracy. This success comes from advances transferred from other machine learning areas, as well as methods specifically designed to deal with protein sequences and structures, and their abstractions. Novel emerging approaches include (i) geometric learning, that is, learning on representations such as graphs, three-dimensional (3D) Voronoi tessellations, and point clouds; (ii) pretrained protein language models leveraging attention; (iii) equivariant architectures preserving the symmetry of 3D space; (iv) use of large meta-genome databases; (v) combinations of protein representations; and (vi) finally truly end-to-end architectures, that is, differentiable models starting from a sequence and returning a 3D structure. Here, we provide an overview and our opinion of the novel deep learning approaches developed in the last 2 years and widely used in CASP14.

OPUS-X: an open-source toolkit for protein torsion angles, secondary structure, solvent accessibility, contact map predictions and 3D folding

MOTIVATION

The development of an open-source platform to predict protein 1D features and 3D structure is an important task. In this paper, we report an open-source toolkit for protein 3D structure modeling, named OPUS-X. It contains three modules: OPUS-TASS2, which predicts protein torsion angles, secondary structure and solvent accessibility; OPUS-Contact, which measures the distance and orientation information between different residue pairs; and OPUS-Fold2, which uses the constraints derived from the first two modules to guide folding.

RESULTS

OPUS-TASS2 is an upgraded version of our previous method OPUS-TASS. OPUS-TASS2 integrates protein global structure information and significantly outperforms OPUS-TASS. OPUS-Contact combines multiple raw co-evolutionary features with protein 1D features predicted by OPUS-TASS2, and delivers better results than the open-source state-of-the-art method trRosetta. OPUS-Fold2 is a complementary version of our previous method OPUS-Fold. OPUS-Fold2 is a gradient-based protein folding framework based on the differentiable energy terms in opposed to OPUS-Fold that is a sampling-based method used to deal with the non-differentiable terms. OPUS-Fold2 exhibits comparable performance to the Rosetta folding protocol in trRosetta when using identical inputs. OPUS-Fold2 is written in Python and TensorFlow2.4, which is user-friendly to any source-code-level modification.

AVAILABILITYAND IMPLEMENTATION

The code and pre-trained models of OPUS-X can be downloaded from https://github.com/OPUS-MaLab/opus_x.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

When homologous sequences meet structural decoys: Accurate contact prediction by tFold in CASP14-(tFold for CASP14 contact prediction)

In this paper, we report our tFold framework's performance on the inter-residue contact prediction task in the 14th Critical Assessment of protein Structure Prediction (CASP14). Our tFold framework seamlessly combines both homologous sequences and structural decoys under an ultra-deep network architecture. Squeeze-excitation and axial attention mechanisms are employed to effectively capture inter-residue interactions. In CASP14, our best predictor achieves 41.78% in the averaged top-L precision for long-range contacts for all the 22 free-modeling (FM) targets, and ranked 1st among all the 60 participating teams. The tFold web server is now freely available at: https://drug.ai.tencent.com/console/en/tfold.

Accurate prediction of inter-protein residue-residue contacts for homo-oligomeric protein complexes

Protein-protein interactions play a fundamental role in all cellular processes. Therefore, determining the structure of protein-protein complexes is crucial to understand their molecular mechanisms and develop drugs targeting the protein-protein interactions. Recently, deep learning has led to a breakthrough in intra-protein contact prediction, achieving an unusual high accuracy in recent Critical Assessment of protein Structure Prediction (CASP) structure prediction challenges. However, due to the limited number of known homologous protein-protein interactions and the challenge to generate joint multiple sequence alignments of two interacting proteins, the advances in inter-protein contact prediction remain limited. Here, we have proposed a deep learning model to predict inter-protein residue-residue contacts across homo-oligomeric protein interfaces, named as DeepHomo. Unlike previous deep learning approaches, we integrated intra-protein distance map and inter-protein docking pattern, in addition to evolutionary coupling, sequence conservation, and physico-chemical information of monomers. DeepHomo was extensively tested on both experimentally determined structures and realistic CASP-Critical Assessment of Predicted Interaction (CAPRI) targets. It was shown that DeepHomo achieved a high precision of >60% for the top predicted contact and outperformed state-of-the-art direct-coupling analysis and machine learning-based approaches. Integrating predicted inter-chain contacts into protein-protein docking significantly improved the docking accuracy on the benchmark dataset of realistic homo-dimeric targets from CASP-CAPRI experiments. DeepHomo is available at http://huanglab.phys.hust.edu.cn/DeepHomo/.