Fold recognition by combining sequence profiles derived from evolution and from depth-dependent structural alignment of fragments

Design, synthesis and preliminary biological evaluation of rivastigmine-INDY hybrids as multitarget ligands against Alzheimer's disease by targeting butyrylcholinesterase and DYRK1A/CLK1 kinases

Based on a multitarget approach implementing rivastigmine-INDY hybrids 1, we identified a set of pseudo-irreversible carbamate-type inhibitors of eqBuChE that, after carbamate transfer at the active site serine residue, released the corresponding INDY analogues 2 endowed with hDYRK1A/hCLK1 kinases inhibitory properties. A SAR study and molecular docking investigation of both series of compounds 1 and 2 revealed that appropriate structural modifications at the carbamate moiety and at the N-appendage of the benzothiazole core led to potent and selective eqBuChE inhibitors with IC50 up to 27 nM and potent hDYRK1A and hCLK1 inhibitors with IC50 up to 106 nM and 17 nM respectively. Pleasingly, identification of the matched pair of compounds 1b/2b with a good balance between inhibition of eqBuChE and hDYRK1A/hCLK1 kinases (IC50 = 68 nM and IC50 = 529/54 nM, respectively) further validated our multitarget approach based on a sequential mechanism of action. In addition, target compound 1b exhibited a suitable ADMET profile, including good brain permeability and high stability in PBS, encouraging further biological investigation as a drug candidate.

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.

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.

Homology Modeling in the Twilight Zone: Improved Accuracy by Sequence Space Analysis

The analysis of the relationship between sequence and structure similarities during the evolution of a protein family has revealed a limit of sequence divergence for which structural conservation can be confidently assumed and homology modeling is reliable. Below this limit, the twilight zone corresponds to sequence divergence for which homology modeling becomes increasingly difficult and requires specific methods. Either with conventional threading methods or with recent deep learning methods, such as AlphaFold, the challenge relies on the identification of a template that shares not only a common ancestor (homology) but also a conserved structure with the query. As both homology and structural conservation are transitive properties, mining of sequence databases followed by multidimensional scaling (MDS) of the query sequence space can reveal intermediary sequences to infer homology and structural conservation between the query and the template. Here, as a case study, we studied the plethodontid receptivity factor isoform 1 (PRF1) from Plethodon jordani, a member of a pheromone protein family present only in lungless salamanders and weakly related to cytokines of the IL6 family. A variety of conventional threading methods led to the cytokine CNTF as a template. Sequence mining, followed by phylogenetic and MDS analysis, provided missing links between PRF1 and CNTF and allowed reliable homology modeling. In addition, we compared automated models obtained from web servers to a customized model to show how modeling can be improved by expert information.

CRFalign: A Sequence-Structure Alignment of Proteins Based on a Combination of HMM-HMM Comparison and Conditional Random Fields

Sequence–structure alignment for protein sequences is an important task for the template-based modeling of 3D structures of proteins. Building a reliable sequence–structure alignment is a challenging problem, especially for remote homologue target proteins. We built a method of sequence–structure alignment called CRFalign, which improves upon a base alignment model based on HMM-HMM comparison by employing pairwise conditional random fields in combination with nonlinear scoring functions of structural and sequence features. Nonlinear scoring part is implemented by a set of gradient boosted regression trees. In addition to sequence profile features, various position-dependent structural features are employed including secondary structures and solvent accessibilities. Training is performed on reference alignments at superfamily levels or twilight zone chosen from the SABmark benchmark set. We found that CRFalign method produces relative improvement in terms of average alignment accuracies for validation sets of SABmark benchmark. We also tested CRFalign on 51 sequence–structure pairs involving 15 FM target domains of CASP14, where we could see that CRFalign leads to an improvement in average modeling accuracies in these hard targets (TM-CRFalign ≃42.94%) compared with that of HHalign (TM-HHalign ≃39.05%) and also that of MRFalign (TM-MRFalign ≃36.93%). CRFalign was incorporated to our template search framework called CRFpred and was tested for a random target set of 300 target proteins consisting of Easy, Medium and Hard sets which showed a reasonable template search performance.

SelfAT-Fold: Protein Fold Recognition Based on Residue-Based and Motif-Based Self-Attention Networks

The protein fold recognition is a fundamental and crucial step of tertiary structure determination. In this regard, several computational predictors have been proposed. Recently, the predictive performance has been obviously improved by the fold-specific features generated by deep learning techniques. However, these methods failed to measure the global associations among residues or motifs along the protein sequences. Furthermore, these deep learning techniques are often treated as black boxes without interpretability. Inspired by the similarities between protein sequences and natural language sentences, we applied the self-attention mechanism derived from natural language processing (NLP) field to protein fold recognition. The motif-based self-attention network (MSAN) and the residue-based self-attention network (RSAN) were constructed based on a training set to capture the global associations among the structure motifs and residues along the protein sequences, respectively. The fold-specific attention features trained and generated from the training set were then combined with Support Vector Machines (SVMs) to predict the samples in the widely used LE benchmark dataset, which is fully independent from the training set. Experimental results showed that the proposed two SelfAT-Fold predictors outperformed 34 existing state-of-the-art computational predictors. The two SelfAT-Fold predictors were further tested on an independent dataset SCOP_TEST, and they can achieve stable performance. Furthermore, the fold-specific attention features can be used to analyse the characteristics of protein folds. The trained models and data of SelfAT-Fold can be downloaded from http://bliulab.net/selfAT_fold/.

LOMETS3: integrating deep learning and profile alignment for advanced protein template recognition and function annotation

Deep learning techniques have significantly advanced the field of protein structure prediction. LOMETS3 (https://zhanglab.ccmb.med.umich.edu/LOMETS/) is a new generation meta-server approach to template-based protein structure prediction and function annotation, which integrates newly developed deep learning threading methods. For the first time, we have extended LOMETS3 to handle multi-domain proteins and to construct full-length models with gradient-based optimizations. Starting from a FASTA-formatted sequence, LOMETS3 performs four steps of domain boundary prediction, domain-level template identification, full-length template/model assembly and structure-based function prediction. The output of LOMETS3 contains (i) top-ranked templates from LOMETS3 and its component threading programs, (ii) up to 5 full-length structure models constructed by L-BFGS (limited-memory Broyden–Fletcher–Goldfarb–Shanno algorithm) optimization, (iii) the 10 closest Protein Data Bank (PDB) structures to the target, (iv) structure-based functional predictions, (v) domain partition and assembly results, and (vi) the domain-level threading results, including items (i)–(iii) for each identified domain. LOMETS3 was tested in large-scale benchmarks and the blind CASP14 (14th Critical Assessment of Structure Prediction) experiment, where the overall template recognition and function prediction accuracy is significantly beyond its predecessors and other state-of-the-art threading approaches, especially for hard targets without homologous templates in the PDB. Based on the improved developments, LOMETS3 should help significantly advance the capability of broader biomedical community for template-based protein structure and function modelling.

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/.

Protein structure prediction using deep learning distance and hydrogen-bonding restraints in CASP14

In this article, we report 3D structure prediction results by two of our best server groups ("Zhang-Server" and "QUARK") in CASP14. These two servers were built based on the D-I-TASSER and D-QUARK algorithms, which integrated four newly developed components into the classical protein folding pipelines, I-TASSER and QUARK, respectively. The new components include: (a) a new multiple sequence alignment (MSA) collection tool, DeepMSA2, which is extended from the DeepMSA program; (b) a contact-based domain boundary prediction algorithm, FUpred, to detect protein domain boundaries; (c) a residual convolutional neural network-based method, DeepPotential, to predict multiple spatial restraints by co-evolutionary features derived from the MSA; and (d) optimized spatial restraint energy potentials to guide the structure assembly simulations. For 37 FM targets, the average TM-scores of the first models produced by D-I-TASSER and D-QUARK were 96% and 112% higher than those constructed by I-TASSER and QUARK, respectively. The data analysis indicates noticeable improvements produced by each of the four new components, especially for the newly added spatial restraints from DeepPotential and the well-tuned force field that combines spatial restraints, threading templates, and generic knowledge-based potentials. However, challenges still exist in the current pipelines. These include difficulties in modeling multi-domain proteins due to low accuracy in inter-domain distance prediction and modeling protein domains from oligomer complexes, as the co-evolutionary analysis cannot distinguish inter-chain and intra-chain distances. Specifically tuning the deep learning-based predictors for multi-domain targets and protein complexes may be helpful to address these issues.

Folding non-homologous proteins by coupling deep-learning contact maps with I-TASSER assembly simulations

Structure prediction for proteins lacking homologous templates in the Protein Data Bank (PDB) remains a significant unsolved problem. We developed a protocol, C-I-TASSER, to integrate interresidue contact maps from deep neural-network learning with the cutting-edge I-TASSER fragment assembly simulations. Large-scale benchmark tests showed that C-I-TASSER can fold more than twice the number of non-homologous proteins than the I-TASSER, which does not use contacts. When applied to a folding experiment on 8,266 unsolved Pfam families, C-I-TASSER successfully folded 4,162 domain families, including 504 folds that are not found in the PDB. Furthermore, it created correct folds for 85% of proteins in the SARS-CoV-2 genome, despite the quick mutation rate of the virus and sparse sequence profiles. The results demonstrated the critical importance of coupling whole-genome and metagenome-based evolutionary information with optimal structure assembly simulations for solving the problem of non-homologous protein structure prediction.

FoldRec-C2C: protein fold recognition by combining cluster-to-cluster model and protein similarity network

As a key for studying the protein structures, protein fold recognition is playing an important role in predicting the protein structures associated with COVID-19 and other important structures. However, the existing computational predictors only focus on the protein pairwise similarity or the similarity between two groups of proteins from 2-folds. However, the homology relationship among proteins is in a hierarchical structure. The global protein similarity network will contribute to the performance improvement. In this study, we proposed a predictor called FoldRec-C2C to globally incorporate the interactions among proteins into the prediction. For the FoldRec-C2C predictor, protein fold recognition problem is treated as an information retrieval task in nature language processing. The initial ranking results were generated by a surprised ranking algorithm Learning to Rank, and then three re-ranking algorithms were performed on the ranking lists to adjust the results globally based on the protein similarity network, including seq-to-seq model, seq-to-cluster model and cluster-to-cluster model (C2C). When tested on a widely used and rigorous benchmark dataset LINDAHL dataset, FoldRec-C2C outperforms other 34 state-of-the-art methods in this field. The source code and data of FoldRec-C2C can be downloaded from http://bliulab.net/FoldRec-C2C/download.

Recent Advances in Protein Homology Detection Propelled by Inter-Residue Interaction Map Threading

Sequence-based protein homology detection has emerged as one of the most sensitive and accurate approaches to protein structure prediction. Despite the success, homology detection remains very challenging for weakly homologous proteins with divergent evolutionary profile. Very recently, deep neural network architectures have shown promising progress in mining the coevolutionary signal encoded in multiple sequence alignments, leading to reasonably accurate estimation of inter-residue interaction maps, which serve as a rich source of additional information for improved homology detection. Here, we summarize the latest developments in protein homology detection driven by inter-residue interaction map threading. We highlight the emerging trends in distant-homology protein threading through the alignment of predicted interaction maps at various granularities ranging from binary contact maps to finer-grained distance and orientation maps as well as their combination. We also discuss some of the current limitations and possible future avenues to further enhance the sensitivity of protein homology detection.

Development of Novel Potential Pleiotropic Compounds of Interest in Alzheimer's Disease Treatment through Rigidification Strategy

The development of Multi-Target Directed Ligand is of clear interest for the treatment of multifactorial pathology such as Alzheimer’s disease (AD). In this context, acetylcholinesterase (AChE) inhibitors have been modulated in order to generate novel pleiotropic compounds targeting a second protein of therapeutic interest in AD. Among them, donecopride was the first example of a dual acetylcholinesterase inhibitor and 5-HT₄ receptor agonist. In order to explore the structural diversity around this preclinical candidate we have explored the preparation of novel constrained analogs through late-stage rigidification strategy. A series of phenylpyrazoles was prepared in a late-stage functionalization process and all compounds were evaluated in vitro towards AChE and 5-HTRs. A docking study was performed in order to better explain the observed SAR towards AChE, 5-HT₄R and 5-HT₆R and this study led to the description of novel ligand targeting both AChE and 5-HT₆R.

Protein Structure Modeling with MODELLER

Genome sequencing projects have resulted in a rapid increase in the number of known protein sequences. In contrast, only about one-hundredth of these sequences have been characterized at atomic resolution using experimental structure determination methods. Computational protein structure modeling techniques have the potential to bridge this sequence-structure gap. In the following chapter, we present an example that illustrates the use of MODELLER to construct a comparative model for a protein with unknown structure. Automation of a similar protocol has resulted in models of useful accuracy for domains in more than half of all known protein sequences.

ProMod3-A versatile homology modelling toolbox

Computational methods for protein structure modelling are routinely used to complement experimental structure determination, thus they help to address a broad spectrum of scientific questions in biomedical research. The most accurate methods today are based on homology modelling, i.e. detecting a homologue to the desired target sequence that can be used as a template for modelling. Here we present a versatile open source homology modelling toolbox as foundation for flexible and computationally efficient modelling workflows. ProMod3 is a fully scriptable software platform that can perform all steps required to generate a protein model by homology. Its modular design aims at fast prototyping of novel algorithms and implementing flexible modelling pipelines. Common modelling tasks, such as loop modelling, sidechain modelling or generating a full protein model by homology, are provided as production ready pipelines, forming the starting point for own developments and enhancements. ProMod3 is the central software component of the widely used SWISS-MODEL web-server.

ProtFold-DFG: protein fold recognition by combining Directed Fusion Graph and PageRank algorithm

As one of the most important tasks in protein structure prediction, protein fold recognition has attracted more and more attention. In this regard, some computational predictors have been proposed with the development of machine learning and artificial intelligence techniques. However, these existing computational methods are still suffering from some disadvantages. In this regard, we propose a new network-based predictor called ProtFold-DFG for protein fold recognition. We propose the Directed Fusion Graph (DFG) to fuse the ranking lists generated by different methods, which employs the transitive closure to incorporate more relationships among proteins and uses the KL divergence to calculate the relationship between two proteins so as to improve its generalization ability. Finally, the PageRank algorithm is performed on the DFG to accurately recognize the protein folds by considering the global interactions among proteins in the DFG. Tested on a widely used and rigorous benchmark data set, LINDAHL dataset, experimental results show that the ProtFold-DFG outperforms the other 35 competing methods, indicating that ProtFold-DFG will be a useful method for protein fold recognition. The source code and data of ProtFold-DFG can be downloaded from http://bliulab.net/ProtFold-DFG/download.

LOMETS2: improved meta-threading server for fold-recognition and structure-based function annotation for distant-homology proteins

The LOMETS2 server (https://zhanglab.ccmb.med.umich.edu/LOMETS/) is an online meta-threading server system for template-based protein structure prediction. Although the server has been widely used by the community over the last decade, the previous LOMETS server no longer represents the state-of-the-art due to aging of the algorithms and unsatisfactory performance on distant-homology template identification. An extension of the server built on cutting-edge methods, especially techniques developed since the recent CASP experiments, is urgently needed. In this work, we report the recent advancements of the LOMETS2 server, which comprise a number of major new developments, including (i) new state-of-the-art threading programs, including contact-map-based threading approaches, (ii) deep sequence search-based sequence profile construction and (iii) a new web interface design that incorporates structure-based function annotations. Large-scale benchmark tests demonstrated that the integration of the deep profiles and new threading approaches into LOMETS2 significantly improve its structure modeling quality and template detection, where LOMETS2 detected 176% more templates with TM-scores >0.5 than the previous LOMETS server for Hard targets that lacked homologous templates. Meanwhile, the newly incorporated structure-based function prediction helps extend the usefulness of the online server to the broader biological community.

ProDCoNN: Protein design using a convolutional neural network

Designing protein sequences that fold to a given three-dimensional (3D) structure has long been a challenging problem in computational structural biology with significant theoretical and practical implications. In this study, we first formulated this problem as predicting the residue type given the 3D structural environment around the C _α atom of a residue, which is repeated for each residue of a protein. We designed a nine-layer 3D deep convolutional neural network (CNN) that takes as input a gridded box with the atomic coordinates and types around a residue. Several CNN layers were designed to capture structure information at different scales, such as bond lengths, bond angles, torsion angles, and secondary structures. Trained on a very large number of protein structures, the method, called ProDCoNN (protein design with CNN), achieved state-of-the-art performance when tested on large numbers of test proteins and benchmark datasets.

To Improve Protein Sequence Profile Prediction through Image Captioning on Pairwise Residue Distance Map

Protein sequence profile prediction aims to generate multiple sequences from structural information to advance the protein design. Protein sequence profile can be computationally predicted by energy-based or fragment-based methods. By integrating these methods with neural networks, our previous method, SPIN2, has achieved a sequence recovery rate of 34%. However, SPIN2 employed only one-dimensional (1D) structural properties that are not sufficient to represent three-dimensional (3D) structures. In this study, we represented 3D structures by 2D maps of pairwise residue distances and developed a new method (SPROF) to predict protein sequence profiles based on an image captioning learning frame. To our best knowledge, this is the first method to employ a 2D distance map for predicting protein properties. SPROF achieved 39.8% in sequence recovery of residues on the independent test set, representing a 5.2% improvement over SPIN2. We also found the sequence recovery increased with the number of their neighbored residues in 3D structural space, indicating that our method can effectively learn long-range information from the 2D distance map. Thus, such network architecture using a 2D distance map is expected to be useful for other 3D structure-based applications, such as binding site prediction, protein function prediction, and protein interaction prediction. The online server and the source code is available at http://biomed.nscc-gz.cn and https://github.com/biomed-AI/SPROF , respectively.

On the possible origin of protein homochirality, structure, and biochemical function

Living systems have chiral molecules, e.g., native proteins that almost entirely contain L-amino acids. How protein homochirality emerged from a background of equal numbers of L and D amino acids is among many questions about life's origin. The origin of homochirality and its implications are explored in computer simulations examining the stability and structural and functional properties of an artificial library of compact proteins containing 1:1 (termed demi-chiral), 3:1, and 1:3 ratios of D:L and purely L or D amino acids generated without functional selection. Demi-chiral proteins have shorter secondary structures and fewer internal hydrogen bonds and are less stable than homochiral proteins. Selection for hydrogen bonding yields a preponderance of L or D amino acids. Demi-chiral proteins have native global folds, including similarity to early ribosomal proteins, similar small molecule ligand binding pocket geometries, and many constellations of L-chiral amino acids with a 1.0-Å RMSD to native enzyme active sites. For a representative subset containing 550 active site geometries matching 457 (2) 4-digit (3-digit) enzyme classification (E.C.) numbers, native active site amino acids were generated at random for 472 of 550 cases. This increases to 548 of 550 cases when similar residues are allowed. The most frequently generated sequences correspond to ancient enzymatic functions, e.g., glycolysis, replication, and nucleotide biosynthesis. Surprisingly, even without selection, demi-chiral proteins possess the requisite marginal biochemical function and structure of modern proteins, but were thermodynamically less stable. If demi-chiral proteins were present, they could engage in early metabolism, which created the feedback loop for transcription and cell formation.

Fold-LTR-TCP: protein fold recognition based on triadic closure principle

As an important task in protein structure and function studies, protein fold recognition has attracted more and more attention. The existing computational predictors in this field treat this task as a multi-classification problem, ignoring the relationship among proteins in the dataset. However, previous studies showed that their relationship is critical for protein homology analysis. In this study, the protein fold recognition is treated as an information retrieval task. The Learning to Rank model (LTR) was employed to retrieve the query protein against the template proteins to find the template proteins in the same fold with the query protein in a supervised manner. The triadic closure principle (TCP) was performed on the ranking list generated by the LTR to improve its accuracy by considering the relationship among the query protein and the template proteins in the ranking list. Finally, a predictor called Fold-LTR-TCP was proposed. The rigorous test on the LE benchmark dataset showed that the Fold-LTR-TCP predictor achieved an accuracy of 73.2%, outperforming all the other competing methods.

MotifCNN-fold: protein fold recognition based on fold-specific features extracted by motif-based convolutional neural networks

Protein fold recognition is one of the most critical tasks to explore the structures and functions of the proteins based on their primary sequence information. The existing protein fold recognition approaches rely on features reflecting the characteristics of protein folds. However, the feature extraction methods are still the bottleneck of the performance improvement of these methods. In this paper, we proposed two new feature extraction methods called MotifCNN and MotifDCNN to extract more discriminative fold-specific features based on structural motif kernels to construct the motif-based convolutional neural networks (CNNs). The pairwise sequence similarity scores calculated based on fold-specific features are then fed into support vector machines to construct the predictor for fold recognition, and a predictor called MotifCNN-fold has been proposed. Experimental results on the benchmark dataset showed that MotifCNN-fold obviously outperformed all the other competing methods. In particular, the fold-specific features extracted by MotifCNN and MotifDCNN are more discriminative than the fold-specific features extracted by other deep learning techniques, indicating that incorporating the structural motifs into the CNN is able to capture the characteristics of protein folds.

DeepSVM-fold: protein fold recognition by combining support vector machines and pairwise sequence similarity scores generated by deep learning networks

Protein fold recognition is critical for studying the structures and functions of proteins. The existing protein fold recognition approaches failed to efficiently calculate the pairwise sequence similarity scores of the proteins in the same fold sharing low sequence similarities. Furthermore, the existing feature vectorization strategies are not able to measure the global relationships among proteins from different protein folds. In this article, we proposed a new computational predictor called DeepSVM-fold for protein fold recognition by introducing a new feature vector based on the pairwise sequence similarity scores calculated from the fold-specific features extracted by deep learning networks. The feature vectors are then fed into a support vector machine to construct the predictor. Experimental results on the benchmark dataset (LE) show that DeepSVM-fold obviously outperforms all the other competing methods.

I-TASSER gateway: A protein structure and function prediction server powered by XSEDE

There is an increasing gap between the number of known protein sequences and the number of proteins with experimentally characterized structure and function. To alleviate this issue, we have developed the I-TASSER gateway, an online server for automated and reliable protein structure and function prediction. For a given sequence, I-TASSER starts with template recognition from a known structure library, followed by full-length atomic model construction by iterative assembly simulations of the continuous structural fragments excised from the template alignments. Functional insights are then derived from comparative matching of the predicted model with a library of proteins with known function. The I-TASSER pipeline has been recently integrated with the XSEDE Gateway system to accommodate pressing demand from the user community and increasing computing costs. This report summarizes the configuration of the I-TASSER Gateway with the XSEDE-Comet supercomputer cluster, together with an overview of the I-TASSER method and milestones of its development.

Constructing effective energy functions for protein structure prediction through broadening attraction-basin and reverse Monte Carlo sampling

BACKGROUND

The ab initio approaches to protein structure prediction usually employ the Monte Carlo technique to search the structural conformation that has the lowest energy. However, the widely-used energy functions are usually ineffective for conformation search. How to construct an effective energy function remains a challenging task.

RESULTS

Here, we present a framework to construct effective energy functions for protein structure prediction. Unlike existing energy functions only requiring the native structure to be the lowest one, we attempt to maximize the attraction-basin where the native structure lies in the energy landscape. The underlying rationale is that each energy function determines a specific energy landscape together with a native attraction-basin, and the larger the attraction-basin is, the more likely for the Monte Carlo search procedure to find the native structure. Following this rationale, we constructed effective energy functions as follows: i) To explore the native attraction-basin determined by a certain energy function, we performed reverse Monte Carlo sampling starting from the native structure, identifying the structural conformations on the edge of attraction-basin. ii) To broaden the native attraction-basin, we smoothened the edge points of attraction-basin through tuning weights of energy terms, thus acquiring an improved energy function. Our framework alternates the broadening attraction-basin and reverse sampling steps (thus called BARS) until the native attraction-basin is sufficiently large. We present extensive experimental results to show that using the BARS framework, the constructed energy functions could greatly facilitate protein structure prediction in improving the quality of predicted structures and speeding up conformation search.

CONCLUSION

Using the BARS framework, we constructed effective energy functions for protein structure prediction, which could improve the quality of predicted structures and speed up conformation search as well.

Does inclusion of residue-residue contact information boost protein threading?

Template-based modeling is considered as one of the most successful approaches for protein structure prediction. However, reliably and accurately selecting optimal template proteins from a library of known protein structures having similar folds as the target protein and making correct alignments between the target sequence and the template structures, a template-based modeling technique known as threading, remains challenging, particularly for non- or distantly-homologous protein targets. With the recent advancement in protein residue-residue contact map prediction powered by sequence co-evolution and machine learning, here we systematically analyze the effect of inclusion of residue-residue contact information in improving the accuracy and reliability of protein threading. We develop a new threading algorithm by incorporating various sequential and structural features, and subsequently integrate residue-residue contact information as an additional scoring term for threading template selection. We show that the inclusion of contact information attains statistically significantly better threading performance compared to a baseline threading algorithm that does not utilize contact information when everything else remains the same. Experimental results demonstrate that our contact based threading approach outperforms popular threading method MUSTER, contact-assisted ab initio folding method CONFOLD2, and recent state-of-the-art contact-assisted protein threading methods EigenTHREADER and map_align on several benchmarks. Our study illustrates that the inclusion of contact maps is a promising avenue in protein threading to ultimately help to improve the accuracy of protein structure prediction.

DESTINI: A deep-learning approach to contact-driven protein structure prediction

The amino acid sequence of a protein encodes the blueprint of its native structure. To predict the corresponding structural fold from the protein’s sequence is one of most challenging problems in computational biology. In this work, we introduce DESTINI (deep structural inference for proteins), a novel computational approach that combines a deep-learning algorithm for protein residue/residue contact prediction with template-based structural modelling. For the first time, the significantly improved predictive ability is demonstrated in the large-scale tertiary structure prediction of over 1,200 single-domain proteins. DESTINI successfully predicts the tertiary structure of four times the number of “hard” targets (those with poor quality templates) that were previously intractable, viz, a “glass-ceiling” for previous template-based approaches, and also improves model quality for “easy” targets (those with good quality templates). The significantly better performance by DESTINI is largely due to the incorporation of better contact prediction into template modelling. To understand why deep-learning accomplishes more accurate contact prediction, systematic clustering reveals that deep-learning predicts coherent, native-like contact patterns compared to co-evolutionary analysis. Taken together, this work presents a promising strategy towards solving the protein structure prediction problem.

Using the Predicted Structure of the Amot Coiled Coil Homology Domain to Understand Lipid Binding

Angiomotins (Amots) are a family of adapter proteins that modulate cellular polarity, differentiation, proliferation, and migration. Amot family members have a characteristic lipid-binding domain, the coiled coil homology (ACCH) domain that selectively targets the protein to membranes, which has been directly linked to its regulatory role in the cell. Several spot blot assays were used to validate the regions of the domain that participate in its membrane association, deformation, and vesicle fusion activity, which indicated the need for a structure to define the mechanism. Therefore, we sought to understand the structure-function relationship of this domain in order to find ways to modulate these signaling pathways. After many failed attempts to crystallize the ACCH domain of each Amot family member for structural analysis, we decided to pursue homologous models that could be refined using small angle x-ray scattering data. Theoretical models were produced using the homology software SWISS-MODEL and threading software I-TASSER and LOMETS, followed by comparison to SAXS data for model selection and refinement. We present a theoretical model of the domain that is driven by alpha helices and short random coil regions. These alpha helical regions form a classic dimer interface followed by two wide spread legs that we predict to be the lipid binding interface.

FINDSITEcomb2.0: A New Approach for Virtual Ligand Screening of Proteins and Virtual Target Screening of Biomolecules

Computational approaches for predicting protein-ligand interactions can facilitate drug lead discovery and drug target determination. We have previously developed a threading/structural-based approach, FINDSITEcomb, for the virtual ligand screening of proteins that has been extensively experimentally validated. Even when low resolution predicted protein structures are employed, FINDSITEcomb has the advantage of being faster and more accurate than traditional high-resolution structure-based docking methods. It also overcomes the limitations of traditional QSAR methods that require a known set of seed ligands that bind to the given protein target. Here, we further improve FINDSITEcomb by enhancing its template ligand selection from the PDB/DrugBank/ChEMBL libraries of known protein-ligand interactions by (1) parsing the template proteins and their corresponding binding ligands in the DrugBank and ChEMBL libraries into domains so that the ligands with falsely matched domains to the targets will not be selected as template ligands; (2) applying various thresholds to filter out falsely matched template structures in the structure comparison process and thus their corresponding ligands for template ligand selection. With a sequence identity cutoff of 30% of target to templates and modeled target structures, FINDSITEcomb2.0 is shown to significantly improve upon FINDSITEcomb on the DUD-E benchmark set by increasing the 1% enrichment factor from 16.7 to 22.1, with a p-value of 4.3 × 10-3 by the Student t-test. With an 80% sequence identity cutoff of target to templates for the DUD-E set and modeled target structures, FINDSITEcomb2.0, having a 1% ROC enrichment factor of 52.39, also outperforms state-of-the-art methods that employ machine learning such as a deep convolutional neural network, CNN, with an enrichment of 29.65. Thus, FINDSITEcomb2.0 represents a significant improvement in the state-of-the-art. The FINDSITEcomb2.0 web service is freely available for academic users at http://pwp.gatech.edu/cssb/FINDSITE-COMB-2 .

Recent advances in sequence-based protein structure prediction

The most accurate characterizations of the structure of proteins are provided by structural biology experiments. However, because of the high cost and labor-intensive nature of the structural experiments, the gap between the number of protein sequences and solved structures is widening rapidly. Development of computational methods to accurately model protein structures from sequences is becoming increasingly important to the biological community. In this article, we highlight some important progress in the field of protein structure prediction, especially those related to free modeling (FM) methods that generate structure models without using homologous templates. We also provide a short synopsis of some of the recent advances in FM approaches as demonstrated in the recent Computational Assessment of Structure Prediction competition as well as recent trends and outlook for FM approaches in protein structure prediction.

A Sequential Segment Based Alpha-Helical Transmembrane Protein Alignment Method

Alpha-helical transmembrane protein (αTMP) is one of the two major categories of transmembrane protein (TMP). They are abundant existing in eukaryotic cells and involved in many biological processes. The special physicochemical properties, the structures of αTMP are hard to be experimentally solved, but αTMP's sequential segments are important to determine their conformations, so that TM-specific alignment is necessary to benefit their structure prediction. We used segment information extracted from topology structure and evolutionary information as features to implement a αTMP Segment Alignment method (TMSA). The method was trained using one non-redundant dataset and tested using another non-redundant dataset. Comparing the results to a general alignment method HHalign, TMSA achieved higher alignment accuracy, and easier to recognize the fold of αTMPs.

EigenTHREADER: analogous protein fold recognition by efficient contact map threading

MOTIVATION

Protein fold recognition when appropriate, evolutionarily-related, structural templates can be identified is often trivial and may even be viewed as a solved problem. However in cases where no homologous structural templates can be detected, fold recognition is a notoriously difficult problem ( Moult et al., 2014 ). Here we present EigenTHREADER, a novel fold recognition method capable of identifying folds where no homologous structures can be identified. EigenTHREADER takes a query amino acid sequence, generates a map of intra-residue contacts, and then searches a library of contact maps of known structures. To allow the contact maps to be compared, we use eigenvector decomposition to resolve the principal eigenvectors these can then be aligned using standard dynamic programming algorithms. The approach is similar to the Al-Eigen approach of Di Lena et al. (2010) , but with improvements made both to speed and accuracy. With this search strategy, EigenTHREADER does not depend directly on sequence homology between the target protein and entries in the fold library to generate models. This in turn enables EigenTHREADER to correctly identify analogous folds where little or no sequence homology information is.

RESULTS

EigenTHREADER outperforms well-established fold recognition methods such as pGenTHREADER and HHSearch in terms of True Positive Rate in the difficult task of analogous fold recognition. This should allow template-based modelling to be extended to many new protein families that were previously intractable to homology based fold recognition methods.

AVAILABILITY AND IMPLEMENTATION

All code used to generate these results and the computational protocol can be downloaded from https://github.com/DanBuchan/eigen_scripts . EigenTHREADER, the benchmark code and the data this paper is based on can be downloaded from: http://bioinfadmin.cs.ucl.ac.uk/downloads/eigenTHREADER/ .

CONTACT

d.t.jones@ucl.ac.uk.

Matrix metalloproteinases outside vertebrates

The matrix metalloproteinase (MMP) family belongs to the metzincin clan of zinc-dependent metallopeptidases. Due to their enormous implications in physiology and disease, MMPs have mainly been studied in vertebrates. They are engaged in extracellular protein processing and degradation, and present extensive paralogy, with 23 forms in humans. One characteristic of MMPs is a ~165-residue catalytic domain (CD), which has been structurally studied for 14 MMPs from human, mouse, rat, pig and the oral-microbiome bacterium Tannerella forsythia. These studies revealed close overall coincidence and characteristic structural features, which distinguish MMPs from other metzincins and give rise to a sequence pattern for their identification. Here, we reviewed the literature available on MMPs outside vertebrates and performed database searches for potential MMP CDs in invertebrates, plants, fungi, viruses, protists, archaea and bacteria. These and previous results revealed that MMPs are widely present in several copies in Eumetazoa and higher plants (Tracheophyta), but have just token presence in eukaryotic algae. A few dozen sequences were found in Ascomycota (within fungi) and in double-stranded DNA viruses infecting invertebrates (within viruses). In contrast, a few hundred sequences were found in archaea and >1000 in bacteria, with several copies for some species. Most of the archaeal and bacterial phyla containing potential MMPs are present in human oral and gut microbiomes. Overall, MMP-like sequences are present across all kingdoms of life, but their asymmetric distribution contradicts the vertical descent model from a eubacterial or archaeal ancestor. This article is part of a Special Issue entitled: Matrix Metalloproteinases edited by Rafael Fridman.

Protein Structure Modeling with MODELLER

Genome sequencing projects have resulted in a rapid increase in the number of known protein sequences. In contrast, only about one-hundredth of these sequences have been characterized at atomic resolution using experimental structure determination methods. Computational protein structure modeling techniques have the potential to bridge this sequence-structure gap. In the following chapter, we present an example that illustrates the use of MODELLER to construct a comparative model for a protein with unknown structure. Automation of a similar protocol has resulted in models of useful accuracy for domains in more than half of all known protein sequences.

Comparative Protein Structure Modeling Using MODELLER

Comparative protein structure modeling predicts the three-dimensional structure of a given protein sequence (target) based primarily on its alignment to one or more proteins of known structure (templates). The prediction process consists of fold assignment, target-template alignment, model building, and model evaluation. This unit describes how to calculate comparative models using the program MODELLER and how to use the ModBase database of such models, and discusses all four steps of comparative modeling, frequently observed errors, and some applications. Modeling lactate dehydrogenase from Trichomonas vaginalis (TvLDH) is described as an example. The download and installation of the MODELLER software is also described. © 2016 by John Wiley & Sons, Inc.

Novel small molecule binders of human N-glycanase 1, a key player in the endoplasmic reticulum associated degradation pathway

Peptide:N-glycanase (NGLY1) is an enzyme responsible for cleaving oligosaccharide moieties from misfolded glycoproteins to enable their proper degradation. Deletion and truncation mutations in this gene are responsible for an inherited disorder of the endoplasmic reticulum-associated degradation pathway. However, the literature is unclear whether the disorder is a result of mutations leading to loss-of-function, loss of substrate specificity, loss of protein stability or a combination of these factors. In this communication, without burdening ourselves with the mechanistic underpinning of disease causation because of mutations on the NGLY1 protein, we demonstrate the successful application of virtual ligand screening (VLS) combined with experimental high-throughput validation to the discovery of novel small-molecules that show binding to the transglutaminase domain of NGLY1. Attempts at recombinant expression and purification of six different constructs led to successful expression of five, with three constructs purified to homogeneity. Most mutant variants failed to purify possibly because of misfolding and the resultant exposure of surface hydrophobicity that led to protein aggregation. For the purified constructs, our threading/structure-based VLS algorithm, FINDSITE(comb), was employed to predict ligands that may bind to the protein. Then, the predictions were assessed by high-throughput differential scanning fluorimetry. This led to the identification of nine different ligands that bind to the protein of interest and provide clues to the nature of pharmacophore that facilitates binding. This is the first study that has identified novel ligands that bind to the NGLY1 protein as a possible starting point in the discovery of ligands with potential therapeutic applications in the treatment of the disorder caused by NGLY1 mutants.

Comparative Protein Structure Modeling Using MODELLER

Comparative protein structure modeling predicts the three-dimensional structure of a given protein sequence (target) based primarily on its alignment to one or more proteins of known structure (templates). The prediction process consists of fold assignment, target-template alignment, model building, and model evaluation. This unit describes how to calculate comparative models using the program MODELLER and how to use the ModBase database of such models, and discusses all four steps of comparative modeling, frequently observed errors, and some applications. Modeling lactate dehydrogenase from Trichomonas vaginalis (TvLDH) is described as an example. The download and installation of the MODELLER software is also described. © 2016 by John Wiley & Sons, Inc.

A Stochastic Point Cloud Sampling Method for Multi-Template Protein Comparative Modeling

Generating tertiary structural models for a target protein from the known structure of its homologous template proteins and their pairwise sequence alignment is a key step in protein comparative modeling. Here, we developed a new stochastic point cloud sampling method, called MTMG, for multi-template protein model generation. The method first superposes the backbones of template structures, and the Cα atoms of the superposed templates form a point cloud for each position of a target protein, which are represented by a three-dimensional multivariate normal distribution. MTMG stochastically resamples the positions for Cα atoms of the residues whose positions are uncertain from the distribution, and accepts or rejects new position according to a simulated annealing protocol, which effectively removes atomic clashes commonly encountered in multi-template comparative modeling. We benchmarked MTMG on 1,033 sequence alignments generated for CASP9, CASP10 and CASP11 targets, respectively. Using multiple templates with MTMG improves the GDT-TS score and TM-score of structural models by 2.96–6.37% and 2.42–5.19% on the three datasets over using single templates. MTMG’s performance was comparable to Modeller in terms of GDT-TS score, TM-score, and GDT-HA score, while the average RMSD was improved by a new sampling approach. The MTMG software is freely available at: http://sysbio.rnet.missouri.edu/multicom_toolbox/mtmg.html.

Using iterative fragment assembly and progressive sequence truncation to facilitate phasing and crystal structure determination of distantly related proteins

Molecular replacement (MR) often requires templates with high homology to solve the phase problem in X-ray crystallography. I-TASSER-MR has been developed to test whether the success rate for structure determination of distant-homology proteins could be improved by a combination of iterative fragmental structure-assembly simulations with progressive sequence truncation designed to trim regions with high variation. The pipeline was tested on two independent protein sets consisting of 61 proteins from CASP8 and 100 high-resolution proteins from the PDB. After excluding homologous templates, I-TASSER generated full-length models with an average TM-score of 0.773, which is 12% higher than the best threading templates. Using these as search models, I-TASSER-MR found correct MR solutions for 95 of 161 targets as judged by having a TFZ of >8 or with the final structure closer to the native than the initial search models. The success rate was 16% higher than when using the best threading templates. I-TASSER-MR was also applied to 14 protein targets from structure genomics centers. Seven of these were successfully solved by I-TASSER-MR. These results confirm that advanced structure assembly and progressive structural editing can significantly improve the success rate of MR for targets with distant homology to proteins of known structure.

Improved de novo structure prediction in CASP11 by incorporating coevolution information into Rosetta

We describe CASP11 de novo blind structure predictions made using the Rosetta structure prediction methodology with both automatic and human assisted protocols. Model accuracy was generally improved using coevolution derived residue-residue contact information as restraints during Rosetta conformational sampling and refinement, particularly when the number of sequences in the family was more than three times the length of the protein. The highlight was the human assisted prediction of T0806, a large and topologically complex target with no homologs of known structure, which had unprecedented accuracy-<3.0 Å root-mean-square deviation (RMSD) from the crystal structure over 223 residues. For this target, we increased the amount of conformational sampling over our fully automated method by employing an iterative hybridization protocol. Our results clearly demonstrate, in a blind prediction scenario, that coevolution derived contacts can considerably increase the accuracy of template-free structure modeling. Proteins 2016; 84(Suppl 1):67-75. © 2015 Wiley Periodicals, Inc.

Three-dimensional homology model of GlcNAc-TV glycosyltransferase

The enzyme UDP-N-acetylglucosamine: α-d-mannoside β-1-6 N-acetylglucosaminyltransferase V (GnT-V) catalyzes the transfer of GlcNAc from the UDP-GlcNAc donor to the α-1-6-linked mannose of the trimannosyl core structure of glycoproteins to produce the β-1-6-linked branching of N-linked oligosaccharides. β-1-6-GlcNAc-branched N-glycans are associated with cancer growth and metastasis. Therefore, the inhibition of GnT-V represents a key target for anti-cancer drug development. However, the development of potent and specific inhibitors of GnT-V is hampered by the lack of information on the three-dimensional structure of the enzyme and on the binding characteristics of its substrates. Here we present the first 3D structure of GnT-V as a result of homology modeling. Various alignment methods, docking the donor and acceptor substrates, and molecular dynamics simulation were used to construct seven homology models of GnT-V and characterize the binding of its substrates. The best homology model is consistent with available experimental data. The three-dimensional model, the structure of the enzyme catalytic site and binding information obtained for the donor and acceptor can be useful in studies of the catalytic mechanism and design of inhibitors of GnT-V.

A neural network learning approach for improving the prediction of residue depth based on sequence-derived features

Residue depth is a solvent exposure measure that quantitatively describes the depth of a residue from the protein surface.

Improving Protein Fold Recognition by Deep Learning Networks

For accurate recognition of protein folds, a deep learning network method (DN-Fold) was developed to predict if a given query-template protein pair belongs to the same structural fold. The input used stemmed from the protein sequence and structural features extracted from the protein pair. We evaluated the performance of DN-Fold along with 18 different methods on Lindahl’s benchmark dataset and on a large benchmark set extracted from SCOP 1.75 consisting of about one million protein pairs, at three different levels of fold recognition (i.e., protein family, superfamily, and fold) depending on the evolutionary distance between protein sequences. The correct recognition rate of ensembled DN-Fold for Top 1 predictions is 84.5%, 61.5%, and 33.6% and for Top 5 is 91.2%, 76.5%, and 60.7% at family, superfamily, and fold levels, respectively. We also evaluated the performance of single DN-Fold (DN-FoldS), which showed the comparable results at the level of family and superfamily, compared to ensemble DN-Fold. Finally, we extended the binary classification problem of fold recognition to real-value regression task, which also show a promising performance. DN-Fold is freely available through a web server at http://iris.rnet.missouri.edu/dnfold.

A large-scale conformation sampling and evaluation server for protein tertiary structure prediction and its assessment in CASP11

Background

With more and more protein sequences produced in the genomic era, predicting protein structures from sequences becomes very important for elucidating the molecular details and functions of these proteins for biomedical research. Traditional template-based protein structure prediction methods tend to focus on identifying the best templates, generating the best alignments, and applying the best energy function to rank models, which often cannot achieve the best performance because of the difficulty of obtaining best templates, alignments, and models.

Methods

We developed a large-scale conformation sampling and evaluation method and its servers to improve the reliability and robustness of protein structure prediction. In the first step, our method used a variety of alignment methods to sample relevant and complementary templates and to generate alternative and diverse target-template alignments, used a template and alignment combination protocol to combine alignments, and used template-based and template-free modeling methods to generate a pool of conformations for a target protein. In the second step, it used a large number of protein model quality assessment methods to evaluate and rank the models in the protein model pool, in conjunction with an exception handling strategy to deal with any additional failure in model ranking.

Results

The method was implemented as two protein structure prediction servers: MULTICOM-CONSTRUCT and MULTICOM-CLUSTER that participated in the 11th Critical Assessment of Techniques for Protein Structure Prediction (CASP11) in 2014. The two servers were ranked among the best 10 server predictors.

Conclusions

The good performance of our servers in CASP11 demonstrates the effectiveness and robustness of the large-scale conformation sampling and evaluation. The MULTICOM server is available at: http://sysbio.rnet.missouri.edu/multicom_cluster/.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-015-0775-x) contains supplementary material, which is available to authorized users.