A unified statistical framework for sequence comparison and structure comparison

LoCoHD: a metric for comparing local environments of proteins

Protein folds and the local environments they create can be compared using a variety of differently designed measures, such as the root mean squared deviation, the global distance test, the template modeling score or the local distance difference test. Although these measures have proven to be useful for a variety of tasks, each fails to fully incorporate the valuable chemical information inherent to atoms and residues, and considers these only partially and indirectly. Here, we develop the highly flexible local composition Hellinger distance (LoCoHD) metric, which is based on the chemical composition of local residue environments. Using LoCoHD, we analyze the chemical heterogeneity of amino acid environments and identify valines having the most conserved-, and arginines having the most variable chemical environments. We use LoCoHD to investigate structural ensembles, to evaluate critical assessment of structure prediction (CASP) competitors, to compare the results with the local distance difference test (lDDT) scoring system, and to evaluate a molecular dynamics simulation. We show that LoCoHD measurements provide unique information about protein structures that is distinct from, for example, those derived using the alignment-based RMSD metric, or the similarly distance matrix-based but alignment-free lDDT metric.

Sequence-structure-function relationships in the microbial protein universe

For the past half-century, structural biologists relied on the notion that similar protein sequences give rise to similar structures and functions. While this assumption has driven research to explore certain parts of the protein universe, it disregards spaces that don't rely on this assumption. Here we explore areas of the protein universe where similar protein functions can be achieved by different sequences and different structures. We predict ~200,000 structures for diverse protein sequences from 1,003 representative genomes across the microbial tree of life and annotate them functionally on a per-residue basis. Structure prediction is accomplished using the World Community Grid, a large-scale citizen science initiative. The resulting database of structural models is complementary to the AlphaFold database, with regards to domains of life as well as sequence diversity and sequence length. We identify 148 novel folds and describe examples where we map specific functions to structural motifs. We also show that the structural space is continuous and largely saturated, highlighting the need for a shift in focus across all branches of biology, from obtaining structures to putting them into context and from sequence-based to sequence-structure-function based meta-omics analyses.

Mutational Paths with Sequence-Based Models of Proteins: From Sampling to Mean-Field Characterization

Identifying and characterizing mutational paths is an important issue in evolutionary biology, with potential applications to bioengineering. We here propose an algorithm to sample mutational paths, which we benchmark on exactly solvable models of proteins in silico, and apply to data-driven models of natural proteins learned from sequence data with restricted Boltzmann machines. We then use mean-field theory to characterize paths for different mutational dynamics of interest, and to extend Kimura's estimate of evolutionary distances to sequence-based epistatic models of selection.

Virtual screening of gut microbiome bacteriocins as potential inhibitors of stearoyl-CoA desaturase 1 to regulate adipocyte differentiation and thermogenesis to combat obesity

The gut bacterial strains and their metabolites have been shown to play a significant role in obesity, but the molecular mechanisms underlying this association are largely unresolved. Obesity is a multifactorial problem and is controlled by various mechanisms and pathways to produce and store fat cells. Bacteriocins are secondary metabolites produced by gut bacteria to defend themselves against their competitors. Recently, they have gained great attention due to their role in metabolic disorders, including obesity. Stearoyl-CoA desaturase 1 (SCD1) is a key enzyme involved in the differentiation of adipocytes. The aim of this study is to show the regulation of SCD1 by bacteriocins and thus their importance in obesity control. We screened the human gut bacteriome for the presence of bacteriocins, predicted their structures, and showed their inhibitory role by molecular docking with SCD1. Further, to confirm the docking results, MDS of six top scoring SCD1-bacteriocin complexes were carried out for 100 ns. These six bacteriocins namely, Plantaricin S-beta, Carnolysin, Lactococcin B, Bacteriocin Iic, Plantaricin N, and Thermophilin A, with strong binding affinities, are primarily produced by bacterial strains from the Lactobacillaeacea family. These findings can be the basis of further experiments for enhanced understanding of the underlying mechanisms for obesity control, specifically bacteriocins driven regulation of the SCD1 enzyme. In addition, a consortium of bacterial strains producing these bacteriocins can be developed and used as probiotics for the amelioration of obesity and other metabolic complications.Communicated by Ramaswamy H. Sarma.

Estimating the Similarity between Protein Pockets

With the exponential increase in publicly available protein structures, the comparison of protein binding sites naturally emerged as a scientific topic to explain observations or generate hypotheses for ligand design, notably to predict ligand selectivity for on- and off-targets, explain polypharmacology, and design target-focused libraries. The current review summarizes the state-of-the-art computational methods applied to pocket detection and comparison as well as structural druggability estimates. The major strengths and weaknesses of current pocket descriptors, alignment methods, and similarity search algorithms are presented. Lastly, an exhaustive survey of both retrospective and prospective applications in diverse medicinal chemistry scenarios illustrates the capability of the existing methods and the hurdle that still needs to be overcome for more accurate predictions.

SIMILE enables alignment of tandem mass spectra with statistical significance

Interrelating small molecules according to their aligned fragmentation spectra is central to tandem mass spectrometry-based untargeted metabolomics. Current alignment algorithms do not provide statistical significance and compounds that have multiple delocalized structural differences and therefore often fail to have their fragment ions aligned. Here we align fragmentation spectra with both statistical significance and allowance for multiple chemical differences using Significant Interrelation of MS/MS Ions via Laplacian Embedding (SIMILE). SIMILE yields spectral alignment inferred structural connections in molecular networks that are not found with cosine-based scoring algorithms. In addition, it is now possible to rank spectral alignments based on p-values in the exploration of structural relationships between compounds and enhance the chemical connectivity that can be obtained with molecular networking.

Interrelating metabolites by their fragmentation spectra is central to metabolomics. Here the authors align fragmentation spectra with both statistical significance and allowance for multiple chemical differences using Significant Interrelation of MS/MS Ions via Laplacian Embedding (SIMILE).

Fast protein structure comparison through effective representation learning with contrastive graph neural networks

Protein structure alignment algorithms are often time-consuming, resulting in challenges for large-scale protein structure similarity-based retrieval. There is an urgent need for more efficient structure comparison approaches as the number of protein structures increases rapidly. In this paper, we propose an effective graph-based protein structure representation learning method, GraSR, for fast and accurate structure comparison. In GraSR, a graph is constructed based on the intra-residue distance derived from the tertiary structure. Then, deep graph neural networks (GNNs) with a short-cut connection learn graph representations of the tertiary structures under a contrastive learning framework. To further improve GraSR, a novel dynamic training data partition strategy and length-scaling cosine distance are introduced. We objectively evaluate our method GraSR on SCOPe v2.07 and a new released independent test set from PDB database with a designed comprehensive performance metric. Compared with other state-of-the-art methods, GraSR achieves about 7%-10% improvement on two benchmark datasets. GraSR is also much faster than alignment-based methods. We dig into the model and observe that the superiority of GraSR is mainly brought by the learned discriminative residue-level and global descriptors. The web-server and source code of GraSR are freely available at www.csbio.sjtu.edu.cn/bioinf/GraSR/ for academic use.

The size and shape of protein structures vary considerably. Accurate protein structure comparison usually relies on structure alignment algorithms. However, superimposing two protein structures is relatively time-consuming, which makes it inappropriate for large-scale protein structure retrieval. Alignment-free algorithms are proposed for efficient protein structure comparison over the last few decades. These algorithms first transform the coordinates of atoms in two proteins to fixed-length vectors. Then, the comparison can be done by measuring the distance or similarity between two vectors, which is much faster than alignment. In this study, we propose a novel protein structure representation method for efficient structure comparison. Compared with other state-of-the-art alignment-free methods, our method achieves better performance on both ranking and multi-class classification tasks due to the powerful representation ability of deep graph neural networks. We dig into the model and observe that the superiority of our method is mainly brought by the learned discriminative residue-level and global descriptors.

Structural modeling of Omicron spike protein and its complex with human ACE-2 receptor: Molecular basis for high transmissibility of the virus

Omicron is a new variant of SARS-CoV-2, which is currently infecting people around the world. Spike glycoprotein, an important molecule in pathogenesis of infection has been modeled and the interaction of its Receptor Binding Domain with human ACE-receptor has been analysed by simulation studies. Structural analysis of Omicron spike glycoprotein shows the 30 mutations to be distributed over all domains of the trimeric protein, wherein the mutant residues are seen to be participating in higher number of intra-molecular interactions including two salt bridges emanating from mutant residues thereby stabilizing their conformation, as compared to wild type. Complex of Receptor Binding Domain (RBD) with human ACE-2 receptor shows seven mutations at interacting interface comprising of two ionic interactions, eight hydrogen bonds and seven Van der Waals interactions. The number and quality of these interactions along with other binding biophysical parameters suggests more potency of RBD domain to the receptor as compared to the wild type counterpart. Results of this study explains the high transmissibility of Omicron variant of SARS-CoV-2 that is currently observed across the world.

Computational Study on the Dynamics of Mycobacterium Tuberculosis RNA Polymerase Assembly

Gene regulation is an intricate phenomenon involving precise function of many macromolecular complexes. Molecular basis of this phenomenon is highly complex and cannot be fully understood using a single technique. Computational approaches can play a crucial role in overall understanding of functional and mechanistic features of a protein or an assembly. Large amounts of structural data pertaining to these complexes are publicly available. In this project, we took advantage of the availability of the structural information to unravel functional intricacies of Mycobacterium tuberculosis RNA polymerase upon interaction with RbpA. In this article, we discuss how the knowledge on protein structure and dynamics can be exploited to study function using various computational tools and resources. Overall, this article provides an overview of various computational methods which can be efficiently used to understand the role of any protein. We hope especially the nonexperts in the field could benefit from our article.

InterPepRank: Assessment of Docked Peptide Conformations by a Deep Graph Network

Peptide-protein interactions between a smaller or disordered peptide stretch and a folded receptor make up a large part of all protein-protein interactions. A common approach for modeling such interactions is to exhaustively sample the conformational space by fast-Fourier-transform docking, and then refine a top percentage of decoys. Commonly, methods capable of ranking the decoys for selection fast enough for larger scale studies rely on first-principle energy terms such as electrostatics, Van der Waals forces, or on pre-calculated statistical potentials. We present InterPepRank for peptide-protein complex scoring and ranking. InterPepRank is a machine learning-based method which encodes the structure of the complex as a graph; with physical pairwise interactions as edges and evolutionary and sequence features as nodes. The graph network is trained to predict the LRMSD of decoys by using edge-conditioned graph convolutions on a large set of peptide-protein complex decoys. InterPepRank is tested on a massive independent test set with no targets sharing CATH annotation nor 30% sequence identity with any target in training or validation data. On this set, InterPepRank has a median AUC of 0.86 for finding coarse peptide-protein complexes with LRMSD < 4Å. This is an improvement compared to other state-of-the-art ranking methods that have a median AUC between 0.65 and 0.79. When included as a selection-method for selecting decoys for refinement in a previously established peptide docking pipeline, InterPepRank improves the number of medium and high quality models produced by 80% and 40%, respectively. The InterPepRank program as well as all scripts for reproducing and retraining it are available from: http://wallnerlab.org/InterPepRank.

Structural analysis of COVID-19 spike protein in recognizing the ACE2 receptor of different mammalian species and its susceptibility to viral infection

The pandemic COVID-19 was caused by a novel Coronavirus-2 (SARS-CoV-2) that infects humans through the binding of glycosylated SARS-CoV-2 spike 2 protein to the glycosylated ACE2 receptor. The spike 2 protein recognizes the N-terminal helices of the glycosylated metalloprotease domain in the human ACE2 receptor. To understand the susceptibility of animals for infection and transmission, we did sequence and structure-based molecular interaction analysis of 16 ACE2 receptors from different mammalian species with SARS-CoV-2 spike 2 receptor binding domain. Our comprehensive structure analysis revealed that the natural substitution of amino acid residues Gln24, His34, Phe40, Leu79 and Met82 in the N-terminal α1 and α2 helices of the ACE2 receptor results in loss of crucial network of hydrogen-bonded and hydrophobic interactions with receptor binding domain of SARS-CoV-2 spike protein. Another striking observation is the absence of N-glycosylation site Asn103 in all mammals and many species, lack more than one N-linked glycosylation site in the ACE2 receptor. Based on the loss of crucial interactions and the absence of N-linked glycosylation sites we categorized Felis catus, Equus caballus, Panthera tigris altaica, as highly susceptible while Oryctolagus cuniculus, Bos Tauras, Ovis aries and Capra hircus as moderately susceptible species for infection. Similarly, the E. asinus, Bubalus bubalis, Canis lupus familiaris, Ailuropoda melaleuca and Camelus dromedarius are categorized as low susceptible with Loxodonta Africana, Mus musculus, Sus scrofa and Rattus rattus as least susceptible species for SARS-CoV-2 infection.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1007/s13205-020-02599-2.

Performance of the Alethia CMV Assay for Detection of Cytomegalovirus by Use of Neonatal Saliva Swabs

Congenital cytomegalovirus (cCMV) infection is a major cause of childhood hearing loss and neurodevelopmental delay. Identification of newborns with cCMV infection allows provision of beneficial interventions. However, most infants with cCMV infection have subclinical infection and go undiagnosed. Thus, expanded neonatal CMV testing is increasingly recommended. Saliva is an attractive sample type for CMV testing of newborns, because it is easier to collect than urine and more sensitive for CMV detection than dried blood spots. We evaluated the Alethia CMV assay, a rapid, easy-to-use loop-mediated isothermal amplification method for qualitative detection of CMV DNA in neonatal saliva samples. Saliva swabs were collected prospectively from newborns <21 days old and tested by the Alethia assay according to the manufacturer's instructions. Archived saliva swabs from newborns with cCMV infection were also tested retrospectively. A composite reference method (CRM; two validated PCR assays followed by bidirectional sequencing of amplicons) was performed on all samples as the reference standard comparator. Of 1,480 prospectively collected saliva swabs, 1,472 (99.5%) were negative by both the Alethia assay and CRM, 5 (0.34%) were positive by both the Alethia assay and CRM, and 3 (0.20%) were positive only by the Alethia assay. All 34 (100%) archived swabs from newborns with cCMV infection were positive by both the CRM and the Alethia assay. Overall, the Alethia assay showed 100% and 99.8% positive and negative agreement with the CRM, respectively. The Alethia CMV assay is an accurate method for identifying neonates with cCMV infection and, given its simplicity, appears suitable for CMV testing using neonatal saliva outside a reference laboratory, including remote and resource-limited settings.

SPECS: Integration of side-chain orientation and global distance-based measures for improved evaluation of protein structural models

Significant advancements in the field of protein structure prediction have necessitated the need for objective and robust evaluation of protein structural models by comparing predicted models against the experimentally determined native structures to quantitate their structural similarities. Existing protein model versus native similarity metrics either consider the distances between alpha carbon (Cα) or side-chain atoms for computing the similarity. However, side-chain orientation of a protein plays a critical role in defining its conformation at the atomic-level. Despite its importance, inclusion of side-chain orientation in structural similarity evaluation has not yet been addressed. Here, we present SPECS, a side-chain-orientation-included protein model-native similarity metric for improved evaluation of protein structural models. SPECS combines side-chain orientation and global distance based measures in an integrated framework using the united-residue model of polypeptide conformation for computing model-native similarity. Experimental results demonstrate that SPECS is a reliable measure for evaluating structural similarity at the global level including and beyond the accuracy of Cα positioning. Moreover, SPECS delivers superior performance in capturing local quality aspect compared to popular global Cα positioning-based metrics ranging from models at near-experimental accuracies to models with correct overall folds-making it a robust measure suitable for both high- and moderate-resolution models. Finally, SPECS is sensitive to minute variations in side-chain χ angles even for models with perfect Cα trace, revealing the power of including side-chain orientation. Collectively, SPECS is a versatile evaluation metric covering a wide spectrum of protein modeling scenarios and simultaneously captures complementary aspects of structural similarities at multiple levels of granularities. SPECS is freely available at http://watson.cse.eng.auburn.edu/SPECS/.

Comparative analysis of methods for evaluation of protein models against native structures

MOTIVATION

Measuring discrepancies between protein models and native structures is at the heart of development of protein structure prediction methods and comparison of their performance. A number of different evaluation methods have been developed; however, their comprehensive and unbiased comparison has not been performed.

RESULTS

We carried out a comparative analysis of several popular model assessment methods (RMSD, TM-score, GDT, QCS, CAD-score, LDDT, SphereGrinder and RPF) to reveal their relative strengths and weaknesses. The analysis, performed on a large and diverse model set derived in the course of three latest community-wide CASP experiments (CASP10-12), had two major directions. First, we looked at general differences between the scores by analyzing distribution, correspondence and correlation of their values as well as differences in selecting best models. Second, we examined the score differences taking into account various structural properties of models (stereochemistry, hydrogen bonds, packing of domains and chain fragments, missing residues, protein length and secondary structure). Our results provide a solid basis for an informed selection of the most appropriate score or combination of scores depending on the task at hand.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Topology independent structural matching discovers novel templates for protein interfaces

Motivation

Protein-protein interactions (PPI) are essential for the function of the cellular machinery. The rapid growth of protein-protein complexes with known 3D structures offers a unique opportunity to study PPI to gain crucial insights into protein function and the causes of many diseases. In particular, it would be extremely useful to compare interaction surfaces of monomers, as this would enable the pinpointing of potential interaction surfaces based solely on the monomer structure, without the need to predict the complete complex structure. While there are many structural alignment algorithms for individual proteins, very few have been developed for protein interfaces, and none that can align only the interface residues to other interfaces or surfaces of interacting monomer subunits in a topology independent (non-sequential) manner.

Results

We present InterComp, a method for topology and sequence-order independent structural comparisons. The method is general and can be applied to various structural comparison applications. By representing residues as independent points in space rather than as a sequence of residues, InterComp can be applied to a wide range of problems including interface-surface comparisons and interface-interface comparisons. We demonstrate a use-case by applying InterComp to find similar protein interfaces on the surface of proteins. We show that InterComp pinpoints the correct interface for almost half of the targets (283 of 586) when considering the top 10 hits, and for 24% of the top 1, even when no templates can be found with regular sequence-order dependent structural alignment methods.

Availability and implementation

The source code and the datasets are available at: http://wallnerlab.org/InterComp.

Supplementary information

Supplementary data are available at Bioinformatics online.

Estimation of model accuracy in CASP13

Methods to reliably estimate the accuracy of 3D models of proteins are both a fundamental part of most protein folding pipelines and important for reliable identification of the best models when multiple pipelines are used. Here, we describe the progress made from CASP12 to CASP13 in the field of estimation of model accuracy (EMA) as seen from the progress of the most successful methods in CASP13. We show small but clear progress, that is, several methods perform better than the best methods from CASP12 when tested on CASP13 EMA targets. Some progress is driven by applying deep learning and residue-residue contacts to model accuracy prediction. We show that the best EMA methods select better models than the best servers in CASP13, but that there exists a great potential to improve this further. Also, according to the evaluation criteria based on local similarities, such as lDDT and CAD, it is now clear that single model accuracy methods perform relatively better than consensus-based methods.

Improved protein model quality assessments by changing the target function

Protein modeling quality is an important part of protein structure prediction. We have for more than a decade developed a set of methods for this problem. We have used various types of description of the protein and different machine learning methodologies. However, common to all these methods has been the target function used for training. The target function in ProQ describes the local quality of a residue in a protein model. In all versions of ProQ the target function has been the S-score. However, other quality estimation functions also exist, which can be divided into superposition- and contact-based methods. The superposition-based methods, such as S-score, are based on a rigid body superposition of a protein model and the native structure, while the contact-based methods compare the local environment of each residue. Here, we examine the effects of retraining our latest predictor, ProQ3D, using identical inputs but different target functions. We find that the contact-based methods are easier to predict and that predictors trained on these measures provide some advantages when it comes to identifying the best model. One possible reason for this is that contact based methods are better at estimating the quality of multi-domain targets. However, training on the S-score gives the best correlation with the GDT_TS score, which is commonly used in CASP to score the global model quality. To take the advantage of both of these features we provide an updated version of ProQ3D that predicts local and global model quality estimates based on different quality estimates.

How large B-factors can be in protein crystal structures

BACKGROUND

Protein crystal structures are potentially over-interpreted since they are routinely refined without any restraint on the upper limit of atomic B-factors. Consequently, some of their atoms, undetected in the electron density maps, are allowed to reach extremely large B-factors, even above 100 square Angstroms, and their final positions are purely speculative and not based on any experimental evidence.

RESULTS

A strategy to define B-factors upper limits is described here, based on the analysis of protein crystal structures deposited in the Protein Data Bank prior 2008, when the tendency to allow B-factor to arbitrary inflate was limited. This B-factor upper limit (B_max) is determined by extrapolating the relationship between crystal structure average B-factor and percentage of crystal volume occupied by solvent (pcVol) to pcVol =100%, when, ab absurdo, the crystal contains only liquid solvent, the structure of which is, by definition, undetectable in electron density maps.

CONCLUSIONS

It is thus possible to highlight structures with average B-factors larger than B_max, which should be considered with caution by the users of the information deposited in the Protein Data Bank, in order to avoid scientifically deleterious over-interpretations.

A comparative analysis of the foamy and ortho virus capsid structures reveals an ancient domain duplication

BACKGROUND

The Spumaretrovirinae (foamy viruses) and the Orthoretrovirinae (e.g. HIV) share many similarities both in genome structure and the sequences of the core viral encoded proteins, such as the aspartyl protease and reverse transcriptase. Similarity in the gag region of the genome is less obvious at the sequence level but has been illuminated by the recent solution of the foamy virus capsid (CA) structure. This revealed a clear structural similarity to the orthoretrovirus capsids but with marked differences that left uncertainty in the relationship between the two domains that comprise the structure.

METHODS

We have applied protein structure comparison methods in order to try and resolve this ambiguous relationship. These included both the DALI method and the SAP method, with rigorous statistical tests applied to the results of both methods. For this, we employed collections of artificial fold 'decoys' (generated from the pair of native structures being compared) to provide a customised background distribution for each comparison, thus allowing significance levels to be estimated.

RESULTS

We have shown that the relationship of the two domains conforms to a simple linear correspondence rather than a domain transposition. These similarities suggest that the origin of both viral capsids was a common ancestor with a double domain structure. In addition, we show that there is also a significant structural similarity between the amino and carboxy domains in both the foamy and ortho viruses.

CONCLUSIONS

These results indicate that, as well as the duplication of the double domain capsid, there may have been an even more ancient gene-duplication that preceded the double domain structure. In addition, our structure comparison methodology demonstrates a general approach to problems where the components have a high intrinsic level of similarity.

RNA Structure: Advances and Assessment of 3D Structure Prediction

Biological functions of RNA molecules are dependent upon sustained specific three-dimensional (3D) structures of RNA, with or without the help of proteins. Understanding of RNA structure is frequently based on 2D structures, which describe only the Watson-Crick (WC) base pairs. Here, we hierarchically review the structural elements of RNA and how they contribute to RNA 3D structure. We focus our analysis on the non-WC base pairs and on RNA modules. Several computer programs have now been designed to predict RNA modules. We describe the RNA-Puzzles initiative, which is a community-wide, blind assessment of RNA 3D structure prediction programs to determine the capabilities and bottlenecks of current predictions. The assessment metrics used in RNA-Puzzles are briefly described. The detection of RNA 3D modules from sequence data and their automatic implementation belong to the current challenges in RNA 3D structure prediction.

Identify High-Quality Protein Structural Models by Enhanced K-Means

Background. One critical issue in protein three-dimensional structure prediction using either ab initio or comparative modeling involves identification of high-quality protein structural models from generated decoys. Currently, clustering algorithms are widely used to identify near-native models; however, their performance is dependent upon different conformational decoys, and, for some algorithms, the accuracy declines when the decoy population increases. Results. Here, we proposed two enhanced K-means clustering algorithms capable of robustly identifying high-quality protein structural models. The first one employs the clustering algorithm SPICKER to determine the initial centroids for basic K-means clustering (SK-means), whereas the other employs squared distance to optimize the initial centroids (K-means++). Our results showed that SK-means and K-means++ were more robust as compared with SPICKER alone, detecting 33 (59%) and 42 (75%) of 56 targets, respectively, with template modeling scores better than or equal to those of SPICKER. Conclusions. We observed that the classic K-means algorithm showed a similar performance to that of SPICKER, which is a widely used algorithm for protein-structure identification. Both SK-means and K-means++ demonstrated substantial improvements relative to results from SPICKER and classical K-means.

Prediction of HIV-1 protease cleavage site using a combination of sequence, structural, and physicochemical features

Background

The human immunodeficiency virus type 1 (HIV-1) aspartic protease is an important enzyme owing to its imperative part in viral development and a causative agent of deadliest disease known as acquired immune deficiency syndrome (AIDS). Development of HIV-1 protease inhibitors can help understand the specificity of substrates which can restrain the replication of HIV-1, thus antagonize AIDS. However, experimental methods in identification of HIV-1 protease cleavage sites are generally time-consuming and labor-intensive. Therefore, using computational methods to predict cleavage sites has become highly desirable.

Results

In this study, we propose a prediction method in which sequence, structural, and physicochemical features are incorporated in various machine learning algorithms. Then, a bidirectional stepwise selection algorithm is incorporated in feature selection to identify discriminative features. Further, only the selected features are calculated by various encoding schemes and used as input for decision trees, logistic regression, and artificial neural networks. Moreover, a more rigorous three-way data split procedure is applied to evaluate the objective performance of cleavage site prediction. Four benchmark datasets collected from previous studies are used to evaluate the predictive performance.

Conclusions

Experiment results showed that combinations of sequence, structure, and physicochemical features performed better than single feature type for identification of HIV-1 protease cleavage sites. In addition, incorporation of stepwise feature selection is effective to identify interpretable biological features to depict specificity of the substrates. Moreover, artificial neural networks perform significantly better than the other two classifiers. Finally, the proposed method achieved 80.0% ~ 97.4% in accuracy and 0.815 ~ 0.995 evaluated by independent test sets in a three-way data split procedure.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-016-1337-6) contains supplementary material, which is available to authorized users.

Parallel-SymD: A Parallel Approach to Detect Internal Symmetry in Protein Domains

Internally symmetric proteins are proteins that have a symmetrical structure in their monomeric single-chain form. Around 10–15% of the protein domains can be regarded as having some sort of internal symmetry. In this regard, we previously published SymD (symmetry detection), an algorithm that determines whether a given protein structure has internal symmetry by attempting to align the protein to its own copy after the copy is circularly permuted by all possible numbers of residues. SymD has proven to be a useful algorithm to detect symmetry. In this paper, we present a new parallelized algorithm called Parallel-SymD for detecting symmetry of proteins on clusters of computers. The achieved speedup of the new Parallel-SymD algorithm scales well with the number of computing processors. Scaling is better for proteins with a larger number of residues. For a protein of 509 residues, a speedup of 63 was achieved on a parallel system with 100 processors.

An efficient algorithm for protein structure comparison using elastic shape analysis

BACKGROUND

Protein structure comparison play important role in in silico functional prediction of a new protein. It is also used for understanding the evolutionary relationships among proteins. A variety of methods have been proposed in literature for comparing protein structures but they have their own limitations in terms of accuracy and complexity with respect to computational time and space. There is a need to improve the computational complexity in comparison/alignment of proteins through incorporation of important biological and structural properties in the existing techniques.

RESULTS

An efficient algorithm has been developed for comparing protein structures using elastic shape analysis in which the sequence of 3D coordinates atoms of protein structures supplemented by additional auxiliary information from side-chain properties are incorporated. The protein structure is represented by a special function called square-root velocity function. Furthermore, singular value decomposition and dynamic programming have been employed for optimal rotation and optimal matching of the proteins, respectively. Also, geodesic distance has been calculated and used as the dissimilarity score between two protein structures. The performance of the developed algorithm is tested and found to be more efficient, i.e., running time reduced by 80-90 % without compromising accuracy of comparison when compared with the existing methods. Source codes for different functions have been developed in R. Also, user friendly web-based application called ProtSComp has been developed using above algorithm for comparing protein 3D structures and is accessible free.

CONCLUSIONS

The methodology and algorithm developed in this study is taking considerably less computational time without loss of accuracy (Table 2). The proposed algorithm is considering different criteria of representing protein structures using 3D coordinates of atoms and inclusion of residue wise molecular properties as auxiliary information.

A Systematic Analysis of the Structures of Heterologously Expressed Proteins and Those from Their Native Hosts in the RCSB PDB Archive

Recombinant expression of proteins has become an indispensable tool in modern day research. The large yields of recombinantly expressed proteins accelerate the structural and functional characterization of proteins. Nevertheless, there are literature reported that the recombinant proteins show some differences in structure and function as compared with the native ones. Now there have been more than 100,000 structures (from both recombinant and native sources) publicly available in the Protein Data Bank (PDB) archive, which makes it possible to investigate if there exist any proteins in the RCSB PDB archive that have identical sequence but have some difference in structures. In this paper, we present the results of a systematic comparative study of the 3D structures of identical naturally purified versus recombinantly expressed proteins. The structural data and sequence information of the proteins were mined from the RCSB PDB archive. The combinatorial extension (CE), FATCAT-flexible and TM-Align methods were employed to align the protein structures. The root-mean-square distance (RMSD), TM-score, P-value, Z-score, secondary structural elements and hydrogen bonds were used to assess the structure similarity. A thorough analysis of the PDB archive generated five-hundred-seventeen pairs of native and recombinant proteins that have identical sequence. There were no pairs of proteins that had the same sequence and significantly different structural fold, which support the hypothesis that expression in a heterologous host usually could fold correctly into their native forms.

A comparison of different functions for predicted protein model quality assessment

In protein structure prediction, a considerable number of models are usually produced by either the Template-Based Method (TBM) or the ab initio prediction. The purpose of this study is to find the critical parameter in assessing the quality of the predicted models. A non-redundant template library was developed and 138 target sequences were modeled. The target sequences were all distant from the proteins in the template library and were aligned with template library proteins on the basis of the transformation matrix. The quality of each model was first assessed with QMEAN and its six parameters, which are C_β interaction energy (C_beta), all-atom pairwise energy (PE), solvation energy (SE), torsion angle energy (TAE), secondary structure agreement (SSA), and solvent accessibility agreement (SAE). Finally, the alignment score (score) was also used to assess the quality of model. Hence, a total of eight parameters (i.e., QMEAN, C_beta, PE, SE, TAE, SSA, SAE, score) were independently used to assess the quality of each model. The results indicate that SSA is the best parameter to estimate the quality of the model.

General overview on structure prediction of twilight-zone proteins

Protein structure prediction from amino acid sequence has been one of the most challenging aspects in computational structural biology despite significant progress in recent years showed by critical assessment of protein structure prediction (CASP) experiments. When experimentally determined structures are unavailable, the predictive structures may serve as starting points to study a protein. If the target protein consists of homologous region, high-resolution (typically <1.5 Å) model can be built via comparative modelling. However, when confronted with low sequence similarity of the target protein (also known as twilight-zone protein, sequence identity with available templates is less than 30%), the protein structure prediction has to be initiated from scratch. Traditionally, twilight-zone proteins can be predicted via threading or ab initio method. Based on the current trend, combination of different methods brings an improved success in the prediction of twilight-zone proteins. In this mini review, the methods, progresses and challenges for the prediction of twilight-zone proteins were discussed.

UniAlign: protein structure alignment meets evolution

MOTIVATION

During the evolution, functional sites on the surface of the protein as well as the hydrophobic core maintaining the structural integrity are well-conserved. However, available protein structure alignment methods align protein structures based solely on the 3D geometric similarity, limiting their ability to detect functionally relevant correspondences between the residues of the proteins, especially for distantly related homologous proteins.

RESULTS

In this article, we propose a new protein pairwise structure alignment algorithm (UniAlign) that incorporates additional evolutionary information captured in the form of sequence similarity, sequence profiles and residue conservation. We define a per-residue score (UniScore) as a weighted sum of these and other features and develop an iterative optimization procedure to search for an alignment with the best overall UniScore. Our extensive experiments on CDD, HOMSTRAD and BAliBASE benchmark datasets show that UniAlign outperforms commonly used structure alignment methods. We further demonstrate UniAlign's ability to develop family-specific models to drastically improve the quality of the alignments.

AVAILABILITY AND IMPLEMENTATION

UniAlign is available as a web service at: http://sacan.biomed.drexel.edu/unialign

CONTACT

ahmet.sacan@drexel.edu

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Geometrical comparison of two protein structures using Wigner-D functions

In this article, we develop a quantitative comparison method for two arbitrary protein structures. This method uses a root-mean-square deviation characterization and employs a series expansion of the protein's shape function in terms of the Wigner-D functions to define a new criterion, which is called a "similarity value." We further demonstrate that the expansion coefficients for the shape function obtained with the help of the Wigner-D functions correspond to structure factors. Our method addresses the common problem of comparing two proteins with different numbers of atoms. We illustrate it with a worked example.

Exploring representations of protein structure for automated remote homology detection and mapping of protein structure space

BACKGROUND

Due to rapid sequencing of genomes, there are now millions of deposited protein sequences with no known function. Fast sequence-based comparisons allow detecting close homologs for a protein of interest to transfer functional information from the homologs to the given protein. Sequence-based comparison cannot detect remote homologs, in which evolution has adjusted the sequence while largely preserving structure. Structure-based comparisons can detect remote homologs but most methods for doing so are too expensive to apply at a large scale over structural databases of proteins. Recently, fragment-based structural representations have been proposed that allow fast detection of remote homologs with reasonable accuracy. These representations have also been used to obtain linearly-reducible maps of protein structure space. It has been shown, as additionally supported from analysis in this paper that such maps preserve functional co-localization of the protein structure space.

METHODS

Inspired by a recent application of the Latent Dirichlet Allocation (LDA) model for conducting structural comparisons of proteins, we propose higher-order LDA-obtained topic-based representations of protein structures to provide an alternative route for remote homology detection and organization of the protein structure space in few dimensions. Various techniques based on natural language processing are proposed and employed to aid the analysis of topics in the protein structure domain.

RESULTS

We show that a topic-based representation is just as effective as a fragment-based one at automated detection of remote homologs and organization of protein structure space. We conduct a detailed analysis of the information content in the topic-based representation, showing that topics have semantic meaning. The fragment-based and topic-based representations are also shown to allow prediction of superfamily membership.

CONCLUSIONS

This work opens exciting venues in designing novel representations to extract information about protein structures, as well as organizing and mining protein structure space with mature text mining tools.

SMOQ: a tool for predicting the absolute residue-specific quality of a single protein model with support vector machines

Background

It is important to predict the quality of a protein structural model before its native structure is known. The method that can predict the absolute local quality of individual residues in a single protein model is rare, yet particularly needed for using, ranking and refining protein models.

Results

We developed a machine learning tool (SMOQ) that can predict the distance deviation of each residue in a single protein model. SMOQ uses support vector machines (SVM) with protein sequence and structural features (i.e. basic feature set), including amino acid sequence, secondary structures, solvent accessibilities, and residue-residue contacts to make predictions. We also trained a SVM model with two new additional features (profiles and SOV scores) on 20 CASP8 targets and found that including them can only improve the performance when real deviations between native and model are higher than 5Å. The SMOQ tool finally released uses the basic feature set trained on 85 CASP8 targets. Moreover, SMOQ implemented a way to convert predicted local quality scores into a global quality score. SMOQ was tested on the 84 CASP9 single-domain targets. The average difference between the residue-specific distance deviation predicted by our method and the actual distance deviation on the test data is 2.637Å. The global quality prediction accuracy of the tool is comparable to other good tools on the same benchmark.

Conclusion

SMOQ is a useful tool for protein single model quality assessment. Its source code and executable are available at: http://sysbio.rnet.missouri.edu/multicom_toolbox/.

Assessment of the assessment: evaluation of the model quality estimates in CASP10

The article presents an assessment of the ability of the thirty-seven model quality assessment (MQA) methods participating in CASP10 to provide an a priori estimation of the quality of structural models, and of the 67 tertiary structure prediction groups to provide confidence estimates for their predicted coordinates. The assessment of MQA predictors is based on the methods used in previous CASPs, such as correlation between the predicted and observed quality of the models (both at the global and local levels), accuracy of methods in distinguishing between good and bad models as well as good and bad regions within them, and ability to identify the best models in the decoy sets. Several numerical evaluations were used in our analysis for the first time, such as comparison of global and local quality predictors with reference (baseline) predictors and a ROC analysis of the predictors' ability to differentiate between the well and poorly modeled regions. For the evaluation of the reliability of self-assessment of the coordinate errors, we used the correlation between the predicted and observed deviations of the coordinates and a ROC analysis of correctly identified errors in the models. A modified two-stage procedure for testing MQA methods in CASP10 whereby a small number of models spanning the whole range of model accuracy was released first followed by the release of a larger number of models of more uniform quality, allowed a more thorough analysis of abilities and inabilities of different types of methods. Clustering methods were shown to have an advantage over the single- and quasi-single- model methods on the larger datasets. At the same time, the evaluation revealed that the size of the dataset has smaller influence on the global quality assessment scores (for both clustering and nonclustering methods), than its diversity. Narrowing the quality range of the assessed models caused significant decrease in accuracy of ranking for global quality predictors but essentially did not change the results for local predictors. Self-assessment error estimates submitted by the majority of groups were poor overall, with two research groups showing significantly better results than the remaining ones.

BAYESIAN PROTEIN STRUCTURE ALIGNMENT

The analysis of the three-dimensional structure of proteins is an important topic in molecular biochemistry. Structure plays a critical role in defining the function of proteins and is more strongly conserved than amino acid sequence over evolutionary timescales. A key challenge is the identification and evaluation of structural similarity between proteins; such analysis can aid in understanding the role of newly discovered proteins and help elucidate evolutionary relationships between organisms. Computational biologists have developed many clever algorithmic techniques for comparing protein structures, however, all are based on heuristic optimization criteria, making statistical interpretation somewhat difficult. Here we present a fully probabilistic framework for pairwise structural alignment of proteins. Our approach has several advantages, including the ability to capture alignment uncertainty and to estimate key "gap" parameters which critically affect the quality of the alignment. We show that several existing alignment methods arise as maximum a posteriori estimates under specific choices of prior distributions and error models. Our probabilistic framework is also easily extended to incorporate additional information, which we demonstrate by including primary sequence information to generate simultaneous sequence-structure alignments that can resolve ambiguities obtained using structure alone. This combined model also provides a natural approach for the difficult task of estimating evolutionary distance based on structural alignments. The model is illustrated by comparison with well-established methods on several challenging protein alignment examples.

Function-based assessment of structural similarity measurements using metal co-factor orientation

Structure comparison is widely used to quantify protein relationships. Although there are several approaches to calculate structural similarity, specifying significance thresholds for similarity metrics is difficult due to the inherent likeness of common secondary structure elements. In this study, metal co-factor location is used to assess the biological relevance of structural alignments. The distance between the centroids of bound co-factors adds a chemical and function-relevant constraint to the structural superimposition of two proteins. This additional dimension can be used to define cut-off values for discriminating valid and spurious alignments in large alignment sets. The hypothesis underlying our approach is that metal coordination sites constrain structural evolution, thus revealing functional relationships between distantly related proteins. A comparison of three related nitrogenases shows the sequence and fold constraints imposed on the protein structures up to 18 Å away from the centers of their bound metal clusters.

The shape of protein crowders is a major determinant of protein diffusion

As a model for understanding how molecular crowding influences diffusion and transport of proteins in cellular environments, we combined experimental and theoretical approaches to study the diffusion of proteins in highly concentrated protein solutions. Bovine serum albumin and γ-Globulin were chosen as molecular crowders and as tracers. These two proteins are representatives of the main types of plasma protein and have different shapes and sizes. Solutions consisting of one or both proteins were studied. The self-diffusion coefficients of the fluorescently labeled tracer proteins were measured by means of fluorescence correlation spectroscopy at a total protein concentration of up to 400 g/L. γ-Globulin is found to have a stronger influence as a crowder on the tracer self-diffusion coefficient than Bovine serum albumin. Brownian dynamics simulations show that the excluded volume and the shape of the crowding protein have a significantly stronger influence on translational and rotational diffusion coefficients, as well as transient oligomerization, than hydrodynamic or direct interactions. Anomalous subdiffusion, which is not observed at the experimental fluorescence correlation spectroscopy timescales (>100 μs), appears only at very short timescales (<1 μs) in the simulations due to steric effects of the proteins. We envision that the combined experimental and computational approach employed here can be developed to unravel the different biophysical contributions to protein motion and interaction in cellular environments by systematically varying protein properties such as molecular weight, size, shape, and electrostatic interactions.

Homology Models and Molecular Dynamics Simulations of Main Proteinase from Coronavirus Associated with Severe Acute Respiratory Syndrome (SARS)

In this study, two structural models (denoted as M^proST and M^proSH) of the main proteinase (M^pro) from the novel coronavirus associated with severe acute respiratory syndrome (SARS‐CoV) were constructed based on the crystallographic structures of M^pro from transmissible gastroenteritis coronavirus (TGEV) (M^proT) and human coronavirus HcoV‐229E (M^proH), respectively. Various 200 ps molecular dynamics simulations were subsequently performed to investigate the dynamics behaviors of several structural features. Both M^proST and M^proSH exhibit similar folds as their respective template proteins. These structural models reveal three distinct functional domains as well as an intervening loop connecting domains II and III as found in both template proteins. In addition, domain III of these structures exhibits the least secondary structural conservation. A catalytic cleft containing the substrate binding subsites S1 and the S2 between domains I and II are also observed in these structural models. Although these structures share many common features, the most significant difference occurs at the S2 subsite, where the amino acid residues lining up this subsite are least conserved. It may be a critical challenge for designing anti‐SARS drugs by simply screening the known database of proteinase inhibitors.

Protein structure alignment beyond spatial proximity

Protein structure alignment is a fundamental problem in computational structure biology. Many programs have been developed for automatic protein structure alignment, but most of them align two protein structures purely based upon geometric similarity without considering evolutionary and functional relationship. As such, these programs may generate structure alignments which are not very biologically meaningful from the evolutionary perspective. This paper presents a novel method DeepAlign for automatic pairwise protein structure alignment. DeepAlign aligns two protein structures using not only spatial proximity of equivalent residues (after rigid-body superposition), but also evolutionary relationship and hydrogen-bonding similarity. Experimental results show that DeepAlign can generate structure alignments much more consistent with manually-curated alignments than other automatic tools especially when proteins under consideration are remote homologs. These results imply that in addition to geometric similarity, evolutionary information and hydrogen-bonding similarity are essential to aligning two protein structures.

QA-RecombineIt: a server for quality assessment and recombination of protein models

QA-RecombineIt provides a web interface to assess the quality of protein 3D structure models and to improve the accuracy of models by merging fragments of multiple input models. QA-RecombineIt has been developed for protein modelers who are working on difficult problems, have a set of different homology models and/or de novo models (from methods such as I-TASSER or ROSETTA) and would like to obtain one consensus model that incorporates the best parts into one structure that is internally coherent. An advanced mode is also available, in which one can modify the operation of the fragment recombination algorithm by manually identifying individual fragments or entire models to recombine. Our method produces up to 100 models that are expected to be on the average more accurate than the starting models. Therefore, our server may be useful for crystallographic protein structure determination, where protein models are used for Molecular Replacement to solve the phase problem. To address the latter possibility, a special feature was added to the QA-RecombineIt server. The QA-RecombineIt server can be freely accessed at http://iimcb.genesilico.pl/qarecombineit/.

Restricted N-glycan conformational space in the PDB and its implication in glycan structure modeling

Understanding glycan structure and dynamics is central to understanding protein-carbohydrate recognition and its role in protein-protein interactions. Given the difficulties in obtaining the glycan's crystal structure in glycoconjugates due to its flexibility and heterogeneity, computational modeling could play an important role in providing glycosylated protein structure models. To address if glycan structures available in the PDB can be used as templates or fragments for glycan modeling, we present a survey of the N-glycan structures of 35 different sequences in the PDB. Our statistical analysis shows that the N-glycan structures found on homologous glycoproteins are significantly conserved compared to the random background, suggesting that N-glycan chains can be confidently modeled with template glycan structures whose parent glycoproteins share sequence similarity. On the other hand, N-glycan structures found on non-homologous glycoproteins do not show significant global structural similarity. Nonetheless, the internal substructures of these N-glycans, particularly, the substructures that are closer to the protein, show significantly similar structures, suggesting that such substructures can be used as fragments in glycan modeling. Increased interactions with protein might be responsible for the restricted conformational space of N-glycan chains. Our results suggest that structure prediction/modeling of N-glycans of glycoconjugates using structure database could be effective and different modeling approaches would be needed depending on the availability of template structures.

An N-glycan is a carbohydrate chain covalently linked to the side chain of asparagine. Due to the flexibility of carbohydrate chains, it is believed that the N-glycan chains would not have a well-defined structure. However, our survey of N-glycan structures in the PDB shows that the N-glycan structures found on the surfaces of homologous glycoproteins are significantly conserved. This suggests that the interaction between the carbohydrate and the protein structure around the glycan chain plays an important role in determining the N-glycan structure. While the global N-glycan structures found on the surfaces of non-homologous glycoproteins are not conserved, the conformations of the carbohydrate residues that are closer to the protein appear to be more conserved. Our analysis highlights the applicability of template-based approaches used in protein structure prediction to structure prediction and modeling of N-glycans of glycoproteins.

3Drefine: consistent protein structure refinement by optimizing hydrogen bonding network and atomic-level energy minimization

One of the major limitations of computational protein structure prediction is the deviation of predicted models from their experimentally derived true, native structures. The limitations often hinder the possibility of applying computational protein structure prediction methods in biochemical assignment and drug design that are very sensitive to structural details. Refinement of these low-resolution predicted models to high-resolution structures close to the native state, however, has proven to be extremely challenging. Thus, protein structure refinement remains a largely unsolved problem. Critical assessment of techniques for protein structure prediction (CASP) specifically indicated that most predictors participating in the refinement category still did not consistently improve model quality. Here, we propose a two-step refinement protocol, called 3Drefine, to consistently bring the initial model closer to the native structure. The first step is based on optimization of hydrogen bonding (HB) network and the second step applies atomic-level energy minimization on the optimized model using a composite physics and knowledge-based force fields. The approach has been evaluated on the CASP benchmark data and it exhibits consistent improvement over the initial structure in both global and local structural quality measures. 3Drefine method is also computationally inexpensive, consuming only few minutes of CPU time to refine a protein of typical length (300 residues). 3Drefine web server is freely available at http://sysbio.rnet.missouri.edu/3Drefine/.

On the difference in quality between current heuristic and optimal solutions to the protein structure alignment problem

The importance of pairwise protein structural comparison in biomedical research is fueling the search for algorithms capable of finding more accurate structural match of two input proteins in a timely manner. In recent years, we have witnessed rapid advances in the development of methods for approximate and optimal solutions to the protein structure matching problem. Albeit slow, these methods can be extremely useful in assessing the accuracy of more efficient, heuristic algorithms. We utilize a recently developed approximation algorithm for protein structure matching to demonstrate that a deep search of the protein superposition space leads to increased alignment accuracy with respect to many well-established measures of alignment quality. The results of our study suggest that a large and important part of the protein superposition space remains unexplored by current techniques for protein structure alignment.

Error-estimation-guided rebuilding ofde novomodels increases the success rate ofab initiophasing

Recent advancements in computational methods for protein-structure prediction have made it possible to generate the high-qualityde novomodels required forab initiophasing of crystallographic diffraction data using molecular replacement. Despite those encouraging achievements inab initiophasing usingde novomodels, its success is limited only to those targets for which high-qualityde novomodels can be generated. In order to increase the scope of targets to whichab initiophasing withde novomodels can be successfully applied, it is necessary to reduce the errors in thede novomodels that are used as templates for molecular replacement. Here, an approach is introduced that can identify and rebuild the residues with larger errors, which subsequently reduces the overall Cαroot-mean-square deviation (CA-RMSD) from the native protein structure. The error in a predicted model is estimated from the average pairwise geometric distance per residue computed among selected lowest energy coarse-grained models. This score is subsequently employed to guide a rebuilding process that focuses on more error-prone residues in the coarse-grained models. This rebuilding methodology has been tested on ten protein targets that were unsuccessful using previous methods. The average CA-RMSD of the coarse-grained models was improved from 4.93 to 4.06 Å. For those models with CA-RMSD less than 3.0 Å, the average CA-RMSD was improved from 3.38 to 2.60 Å. These rebuilt coarse-grained models were then converted into all-atom models and refined to produce improvedde novomodels for molecular replacement. Seven diffraction data sets were successfully phased using rebuiltde novomodels, indicating the improved quality of these rebuiltde novomodels and the effectiveness of the rebuilding process. Software implementing this method, calledMORPHEUS, can be downloaded from http://www.riken.jp/zhangiru/software.html.

Evolutionarily consistent families in SCOP: sequence, structure and function

Background

SCOP is a hierarchical domain classification system for proteins of known structure. The superfamily level has a clear definition: Protein domains belong to the same superfamily if there is structural, functional and sequence evidence for a common evolutionary ancestor. Superfamilies are sub-classified into families, however, there is not such a clear basis for the family level groupings. Do SCOP families group together domains with sequence similarity, do they group domains with similar structure or by common function? It is these questions we answer, but most importantly, whether each family represents a distinct phylogenetic group within a superfamily.

Results

Several phylogenetic trees were generated for each superfamily: one derived from a multiple sequence alignment, one based on structural distances, and the final two from presence/absence of GO terms or EC numbers assigned to domains. The topologies of the resulting trees and confidence values were compared to the SCOP family classification.

Conclusions

We show that SCOP family groupings are evolutionarily consistent to a very high degree with respect to classical sequence phylogenetics. The trees built from (automatically generated) structural distances correlate well, but are not always consistent with SCOP (hand annotated) groupings. Trees derived from functional data are less consistent with the family level than those from structure or sequence, though the majority still agree. Much of GO and EC annotation applies directly to one family or subset of the family; relatively few terms apply at the superfamily level. Maximum sequence diversity within a family is on average 22% but close to zero for superfamilies.

A new size-independent score for pairwise protein structure alignment and its application to structure classification and nucleic-acid binding prediction

A structure alignment program aligns two structures by optimizing a scoring function that measures structural similarity. It is highly desirable that such scoring function is independent of the sizes of proteins in comparison so that the significance of alignment across different sizes of the protein regions aligned is comparable. Here, we developed a new score called SP-score that fixes the cutoff distance at 4 Å and removed the size dependence using a normalization prefactor. We further built a program called SPalign that optimizes SP-score for structure alignment. SPalign was applied to recognize proteins within the same structure fold and having the same function of DNA or RNA binding. For fold discrimination, SPalign improves sensitivity over TMalign for the chain-level comparison by 12% and over DALI for the domain-level comparison by 13% at the same specificity of 99.6%. The difference between TMalign and SPalign at the chain level is due to the inability of TMalign to detect single domain similarity between multidomain proteins. For recognizing nucleic acid binding proteins, SPalign consistently improves over TMalign by 12% and DALI by 31% in average value of Mathews correlation coefficients for four datasets. SPalign with default setting is 14% faster than TMalign. SPalign is expected to be useful for function prediction and comparing structures with or without domains defined. The source code for SPalign and the server are available at http://sparks.informatics.iupui.edu.