Classifying RNA-binding proteins based on electrostatic properties

Protein structure can provide new insight into the biological function of a protein and can enable the design of better experiments to learn its biological roles. Moreover, deciphering the interactions of a protein with other molecules can contribute to the understanding of the protein's function within cellular processes. In this study, we apply a machine learning approach for classifying RNA-binding proteins based on their three-dimensional structures. The method is based on characterizing unique properties of electrostatic patches on the protein surface. Using an ensemble of general protein features and specific properties extracted from the electrostatic patches, we have trained a support vector machine (SVM) to distinguish RNA-binding proteins from other positively charged proteins that do not bind nucleic acids. Specifically, the method was applied on proteins possessing the RNA recognition motif (RRM) and successfully classified RNA-binding proteins from RRM domains involved in protein-protein interactions. Overall the method achieves 88% accuracy in classifying RNA-binding proteins, yet it cannot distinguish RNA from DNA binding proteins. Nevertheless, by applying a multiclass SVM approach we were able to classify the RNA-binding proteins based on their RNA targets, specifically, whether they bind a ribosomal RNA (rRNA), a transfer RNA (tRNA), or messenger RNA (mRNA). Finally, we present here an innovative approach that does not rely on sequence or structural homology and could be applied to identify novel RNA-binding proteins with unique folds and/or binding motifs.     

qNABpredict: Quick,accurate, and taxonomy-aware sequence-based prediction of content of nucleic acid binding amino acids

Protein sequence-based predictors of nucleic acid (NA)-binding include methods that predict NA-binding proteins and NA-binding residues. The residue-level tools produce more details but suffer high computational cost since they must predict every amino acid in the input sequence and rely on multiple sequence alignments. We propose an alternative approach that predicts content (fraction) of the NA-binding residues, offering more information than the protein-level prediction and much shorter runtime than the residue-level tools. Our first-of-its-kind content predictor, qNABpredict, relies on a small, rationally designed and fast-to-compute feature set that represents relevant characteristics extracted from the input sequence and a well-parametrized support vector regression model. We provide two versions of qNABpredict, a taxonomy-agnostic model that can be used for proteins of unknown taxonomic origin and more accurate taxonomy-aware models that are tailored to specific taxonomic kingdoms: archaea, bacteria, eukaryota, and viruses. Empirical tests on a low-similarity test dataset show that qNABpredict is 100 times faster and generates statistically more accurate content predictions when compared to the content extracted from results produced by the residue-level predictors. We also show that qNABpredict's content predictions can be used to improve results generated by the residue-level predictors. We release qNABpredict as a convenient webserver and source code at http://biomine.cs.vcu.edu/servers/qNABpredict/ . This new tool should be particularly useful to predict details of protein–NA interactions for large protein families and proteomes.     

Predicting nucleic acid binding interfaces from structural models of proteins

The function of DNA‐ and RNA‐binding proteins can be inferred from the characterization and accurate prediction of their binding interfaces. However, the main pitfall of various structure‐based methods for predicting nucleic acid binding function is that they are all limited to a relatively small number of proteins for which high‐resolution three‐dimensional structures are available. In this study, we developed a pipeline for extracting functional electrostatic patches from surfaces of protein structural models, obtained using the I‐TASSER protein structure predictor. The largest positive patches are extracted from the protein surface using the patchfinder algorithm. We show that functional electrostatic patches extracted from an ensemble of structural models highly overlap the patches extracted from high‐resolution structures. Furthermore, by testing our pipeline on a set of 55 known nucleic acid binding proteins for which I‐TASSER produces high‐quality models, we show that the method accurately identifies the nucleic acids binding interface on structural models of proteins. Employing a combined patch approach we show that patches extracted from an ensemble of models better predicts the real nucleic acid binding interfaces compared with patches extracted from independent models. Overall, these results suggest that combining information from a collection of low‐resolution structural models could be a valuable approach for functional annotation. We suggest that our method will be further applicable for predicting other functional surfaces of proteins with unknown structure. Proteins 2012. © 2011 Wiley Periodicals, Inc.     

Using electrostatic potentials to predict DNA-binding sites on DNA-binding proteins

A method to detect DNA-binding sites on the surface of a protein structure is important for functional annotation. This work describes the analysis of residue patches on the surface of DNA-binding proteins and the development of a method of predicting DNA-binding sites using a single feature of these surface patches. Surface patches and the DNA-binding sites were initially analysed for accessibility, electrostatic potential, residue propensity, hydrophobicity and residue conservation. From this, it was observed that the DNA-binding sites were, in general, amongst the top 10% of patches with the largest positive electrostatic scores. This knowledge led to the development of a prediction method in which patches of surface residues were selected such that they excluded residues with negative electrostatic scores. This method was used to make predictions for a data set of 56 non-homologous DNA-binding proteins. Correct predictions made for 68% of the data set.     

Prediction of protein structure from ideal forms

For many years it has been accepted that the sequence of a protein can specify its three-dimensional structure. However, there has been limited progress in explaining how the sequence dictates its fold and no attempt to do this computationally without the use of specific structural data has ever succeeded for any protein larger than 100 residues. We describe a method that can predict complex folds up to almost 200 residues using only basic principles that do not include any elements of sequence homology. The method does not simulate the folding chain but generates many thousands of models based on an idealized representation of structure. Each rough model is scored and the best are refined. On a set of five proteins, the correct fold score well and when tested on a set of larger proteins, the correct fold was ranked highest for some proteins more than 150 residues, with others being close topological variants. All other methods that approach this level of success rely on the use of templates or fragments of known structures. Our method is unique in using a database of ideal models based on general packing rules that, in spirit, is closer to an ab initio approach.     

Structure-based prediction of DNA-binding sites on proteins using the empirical preference of electrostatic potential and the shape of molecular surfaces

Protein-DNA interactions play an essential role in the genetic activities of life. Many structures of protein-DNA complexes are already known, but the common rules on how and where proteins bind to DNA have not emerged. Many attempts have been made to predict protein-DNA interactions using structural information, but the success rate is still about 80%. We analyzed 63 protein-DNA complexes by focusing our attention on the shape of the molecular surface of the protein and DNA, along with the electrostatic potential on the surface, and constructed a new statistical evaluation function to make predictions of DNA interaction sites on protein molecular surfaces. The shape of the molecular surface was described by a combination of local and global average curvature, which are intended to describe the small convex and concave and the large-scale concave curvatures of the protein surface preferentially appearing at DNA-binding sites. Using these structural features, along with the electrostatic potential obtained by solving the Poisson-Boltzmann equation numerically, we have developed prediction schemes with 86% and 96% accuracy for DNA-binding and non-DNA-binding proteins, respectively.     

Identification of protein biochemical functions by similarity search using the molecular surface database eF-site

The identification of protein biochemical functions based on their three-dimensional structures is strongly required in the post-genome-sequencing era. We have developed a new method to identify and predict protein biochemical functions using the similarity information of molecular surface geometries and electrostatic potentials on the surfaces. Our prediction system consists of a similarity search method based on a clique search algorithm and the molecular surface database eF-site (electrostatic surface of functional-site in proteins). Using this system, functional sites similar to those of phosphoenoylpyruvate carboxy kinase were detected in several mononucleotide-binding proteins, which have different folds. We also applied our method to a hypothetical protein, MJ0226 from Methanococcus jannaschii, and detected the mononucleotide binding site from the similarity to other proteins having different folds.     

Exhaustive comparison and classification of ligand‐binding surfaces in proteins

Many proteins function by interacting with other small molecules (ligands). Identification of ligand‐binding sites (LBS) in proteins can therefore help to infer their molecular functions. A comprehensive comparison among local structures of LBSs was previously performed, in order to understand their relationships and to classify their structural motifs. However, similar exhaustive comparison among local surfaces of LBSs (patches) has never been performed, due to computational complexity. To enhance our understanding of LBSs, it is worth performing such comparisons among patches and classifying them based on similarities of their surface configurations and electrostatic potentials. In this study, we first developed a rapid method to compare two patches. We then clustered patches corresponding to the same PDB chemical component identifier for a ligand, and selected a representative patch from each cluster. We subsequently exhaustively as compared the representative patches and clustered them using similarity score, PatSim. Finally, the resultant PatSim scores were compared with similarities of atomic structures of the LBSs and those of the ligand‐binding protein sequences and functions. Consequently, we classified the patches into ～2000 well‐characterized clusters. We found that about 63% of these clusters are used in identical protein folds, although about 25% of the clusters are conserved in distantly related proteins and even in proteins with cross‐fold similarity. Furthermore, we showed that patches with higher PatSim score have potential to be involved in similar biological processes.     

Efficient prediction of nucleic acid binding function from low-resolution protein structures

Structural genomics projects as well as ab initio protein structure prediction methods provide structures of proteins with no sequence or fold similarity to proteins with known functions. These are often low-resolution structures that may only include the positions of C alpha atoms. We present a fast and efficient method to predict DNA-binding proteins from just the amino acid sequences and low-resolution, C alpha-only protein models. The method uses the relative proportions of certain amino acids in the protein sequence, the asymmetry of the spatial distribution of certain other amino acids as well as the dipole moment of the molecule. These quantities are used in a linear formula, with coefficients derived from logistic regression performed on a training set, and DNA-binding is predicted based on whether the result is above a certain threshold. We show that the method is insensitive to errors in the atomic coordinates and provides correct predictions even on inaccurate protein models. We demonstrate that the method is capable of predicting proteins with novel binding site motifs and structures solved in an unbound state. The accuracy of our method is close to another, published method that uses all-atom structures, time-consuming calculations and information on conserved residues.     

A comparison of sequence and structure protein domain families as a basis for structural genomics.

MOTIVATION: Protein families can be defined based on structure or sequence similarity. We wanted to compare two protein family databases, one based on structural and one on sequence similarity, to investigate to what extent they overlap, the similarity in definition of corresponding families, and to create a list of large protein families with unknown structure as a resource for structural genomics. We also wanted to increase the sensitivity of fold assignment by exploiting protein family HMMs. RESULTS: We compared Pfam, a protein family database based on sequence similarity, to Scop, which is based on structural similarity. We found that 70% of the Scop families exist in Pfam while 57% of the Pfam families exist in Scop. Most families that occur in both databases correspond well to each other, but in some cases they are different. Such cases highlight situations in which structure and sequence approaches differ significantly. The comparison enabled us to compile a list of the largest families that do not occur in Scop; these are suitable targets for structure prediction and determination, and may be useful to guide projects in structural genomics. It can be noted that 13 out of the 20 largest protein families without a known structure are likely transmembrane proteins. We also exploited Pfam to increase the sensitivity of detecting homologs of proteins with known structure, by comparing query sequences to Pfam HMMs that correspond to Scop families. For SWISSPROT+TREMBL, this yielded an increase in fold assignment from 31% to 42% compared to using FASTA only. This method assigned a structure to 22% of the proteins in Saccharomyces cerevisiae, 24% in Escherichia coli, and 16% in Methanococcus jannaschii.     

High Resolution Structural Studies of Cro Repressor Protein and Implications for DNA Recognition

Abstract

Cro repressor is a small dimeric protein that binds to specific sites on the DNA of bacteriophage λ. The structure of Cro has been determined and suggests that the protein binds to its sequence-specific sites with a pair of two-fold related α-helices of the protein located within successive major grooves of the DNA.

From the known three-dimensional structure of the repressor, model building and energy refinement have been used to develop a detailed model for the presumed complex between Cro and DNA. Recognition of specific DNA binding sites appears to occur via multiple hydrogen bonds between amino acid side chains of the protein and base pair atoms exposed within the major groove of DNA. The Cro:DNA model is consistent with the calculated electrostatic potential energy surface of the protein.

From a series of amino acid sequence and gene sequence comparisons, it appears that a number of other DNA-binding proteins have an α-helical DNA-binding region similar to that seen in Cro. The apparent sequence homology includes not only DNA-binding proteins from different bacteriophages, but also gene-regulatory proteins from bacteria and yeast. It has also been found that the conformations of part of the presumed DNA-binding regions of Cro repressor, λ repressor and CAP gene activator proteins are strikingly similar. Taken together, these results strongly suggest that a two-helical structural unit occurs in the DNA-binding region of many proteins that regulate gene expression. However, the results to date do not suggest that there is a simple one-to-one recognition code between amino acids and bases.

Crystals have been obtained of complexes of Cro with six-base-pair and nine-basepair DNA oligomers, and X-ray analysis of these co-crystals is in progress.     

Template-based recognition of protein fold within the midnight and twilight zones of protein sequence similarity

Most homologous pairs of proteins have no significant sequence similarity to each other and are not identified by direct sequence comparison or profile-based strategies. However, multiple sequence alignments of low similarity homologues typically reveal a limited number of positions that are well conserved despite diversity of function. It may be inferred that conservation at most of these positions is the result of the importance of the contribution of these amino acids to the folding and stability of the protein. As such, these amino acids and their relative positions may define a structural signature. We demonstrate that extraction of this fold template provides the basis for the sequence database to be searched for patterns consistent with the fold, enabling identification of homologs that are not recognized by global sequence analysis. The fold template method was developed to address the need for a tool that could comprehensively search the midnight and twilight zones of protein sequence similarity without reliance on global statistical significance. Manual implementations of the fold template method were performed on three folds--immunoglobulin, c-lectin and TIM barrel. Following proof of concept of the template method, an automated version of the approach was developed. This automated fold template method was used to develop fold templates for 10 of the more populated folds in the SCOP database. The fold template method developed three-dimensional structural motifs or signatures that were able to return a diverse collection of proteins, while maintaining a low false positive rate. Although the results of the manual fold template method were more comprehensive than the automated fold template method, the diversity of the results from the automated fold template method surpassed those of current methods that rely on statistical significance to infer evolutionary relationships among divergent proteins.     

Armadillo: domain boundary prediction by amino acid composition

The identification and annotation of protein domains provides a critical step in the accurate determination of molecular function. Both computational and experimental methods of protein structure determination may be deterred by large multi-domain proteins or flexible linker regions. Knowledge of domains and their boundaries may reduce the experimental cost of protein structure determination by allowing researchers to work on a set of smaller and possibly more successful alternatives. Current domain prediction methods often rely on sequence similarity to conserved domains and as such are poorly suited to detect domain structure in poorly conserved or orphan proteins. We present here a simple computational method to identify protein domain linkers and their boundaries from sequence information alone. Our domain predictor, Armadillo (http://armadillo.blueprint.org), uses any amino acid index to convert a protein sequence to a smoothed numeric profile from which domains and domain boundaries may be predicted. We derived an amino acid index called the domain linker propensity index (DLI) from the amino acid composition of domain linkers using a non-redundant structure dataset. The index indicates that Pro and Gly show a propensity for linker residues while small hydrophobic residues do not. Armadillo predicts domain linker boundaries from Z-score distributions and obtains 35% sensitivity with DLI in a two-domain, single-linker dataset (within +/-20 residues from linker). The combination of DLI and an entropy-based amino acid index increases the overall Armadillo sensitivity to 56% for two domain proteins. Moreover, Armadillo achieves 37% sensitivity for multi-domain proteins, surpassing most other prediction methods. Armadillo provides a simple, but effective method by which prediction of domain boundaries can be obtained with reasonable sensitivity. Armadillo should prove to be a valuable tool for rapidly delineating protein domains in poorly conserved proteins or those with no sequence neighbors. As a first-line predictor, domain meta-predictors could yield improved results with Armadillo predictions.     

A Threading-Based Method for the Prediction of DNA-Binding Proteins with Application to the Human Genome

Diverse mechanisms for DNA-protein recognition have been elucidated in numerous atomic complex structures from various protein families. These structural data provide an invaluable knowledge base not only for understanding DNA-protein interactions, but also for developing specialized methods that predict the DNA-binding function from protein structure. While such methods are useful, a major limitation is that they require an experimental structure of the target as input. To overcome this obstacle, we develop a threading-based method, DNA-Binding-Domain-Threader (DBD-Threader), for the prediction of DNA-binding domains and associated DNA-binding protein residues. Our method, which uses a template library composed of DNA-protein complex structures, requires only the target protein''s sequence. In our approach, fold similarity and DNA-binding propensity are employed as two functional discriminating properties. In benchmark tests on 179 DNA-binding and 3,797 non-DNA-binding proteins, using templates whose sequence identity is less than 30% to the target, DBD-Threader achieves a sensitivity/precision of 56%/86%. This performance is considerably better than the standard sequence comparison method PSI-BLAST and is comparable to DBD-Hunter, which requires an experimental structure as input. Moreover, for over 70% of predicted DNA-binding domains, the backbone Root Mean Square Deviations (RMSDs) of the top-ranked structural models are within 6.5 Å of their experimental structures, with their associated DNA-binding sites identified at satisfactory accuracy. Additionally, DBD-Threader correctly assigned the SCOP superfamily for most predicted domains. To demonstrate that DBD-Threader is useful for automatic function annotation on a large-scale, DBD-Threader was applied to 18,631 protein sequences from the human genome; 1,654 proteins are predicted to have DNA-binding function. Comparison with existing Gene Ontology (GO) annotations suggests that ∼30% of our predictions are new. Finally, we present some interesting predictions in detail. In particular, it is estimated that ∼20% of classic zinc finger domains play a functional role not related to direct DNA-binding.     

Identification of DNA-binding Proteins Using Structural, Electrostatic and Evolutionary Features

DNA-binding proteins (DBPs) participate in various crucial processes in the life-cycle of the cells, and the identification and characterization of these proteins is of great importance. We present here a random forests classifier for identifying DBPs among proteins with known 3D structures. First, clusters of evolutionarily conserved regions (patches) on the surface of proteins were detected using the PatchFinder algorithm; earlier studies showed that these regions are typically the functionally important regions of proteins. Next, we trained a classifier using features like the electrostatic potential, cluster-based amino acid conservation patterns and the secondary structure content of the patches, as well as features of the whole protein, including its dipole moment. Using 10-fold cross-validation on a dataset of 138 DBPs and 110 proteins that do not bind DNA, the classifier achieved a sensitivity and a specificity of 0.90, which is overall better than the performance of published methods. Furthermore, when we tested five different methods on 11 new DBPs that did not appear in the original dataset, only our method annotated all correctly.The resulting classifier was applied to a collection of 757 proteins of known structure and unknown function. Of these proteins, 218 were predicted to bind DNA, and we anticipate that some of them interact with DNA using new structural motifs. The use of complementary computational tools supports the notion that at least some of them do bind DNA.     

Prediction of operator-binding protein by discriminant analysis

A number of operator-binding proteins contain similar sequence features to Cro and cI repressors of bacteriophage and CAP protein of Escherichia coli, such as conserved amino acids at constant positions. However, these sequence patterns also occur in proteins that are not operator-binding. We use sequence analogy information in conjunction with a pattern recognition algorithm. The functional and structural properties, e.g., distributions of hydrophobicity, hydrophilicity, charged amino acids, electrostatic free energy, and helical structures of protein are also considered. Within the framework of discriminant analysis, we calculate the above variables and search for a better combination of variables. To assess the discriminatory power of these variables, we allocated additional sequences and predict DNA-binding regions of regulatory proteins not included in the training set.     

RosettaSurf—A surface-centric computational design approach

Proteins are typically represented by discrete atomic coordinates providing an accessible framework to describe different conformations. However, in some fields proteins are more accurately represented as near-continuous surfaces, as these are imprinted with geometric (shape) and chemical (electrostatics) features of the underlying protein structure. Protein surfaces are dependent on their chemical composition and, ultimately determine protein function, acting as the interface that engages in interactions with other molecules. In the past, such representations were utilized to compare protein structures on global and local scales and have shed light on functional properties of proteins. Here we describe RosettaSurf, a surface-centric computational design protocol, that focuses on the molecular surface shape and electrostatic properties as means for protein engineering, offering a unique approach for the design of proteins and their functions. The RosettaSurf protocol combines the explicit optimization of molecular surface features with a global scoring function during the sequence design process, diverging from the typical design approaches that rely solely on an energy scoring function. With this computational approach, we attempt to address a fundamental problem in protein design related to the design of functional sites in proteins, even when structurally similar templates are absent in the characterized structural repertoire. Surface-centric design exploits the premise that molecular surfaces are, to a certain extent, independent of the underlying sequence and backbone configuration, meaning that different sequences in different proteins may present similar surfaces. We benchmarked RosettaSurf on various sequence recovery datasets and showcased its design capabilities by generating epitope mimics that were biochemically validated. Overall, our results indicate that the explicit optimization of surface features may lead to new routes for the design of functional proteins.     

Kernel-based machine learning protocol for predicting DNA-binding proteins

DNA-binding proteins (DNA-BPs) play a pivotal role in various intra- and extra-cellular activities ranging from DNA replication to gene expression control. Attempts have been made to identify DNA-BPs based on their sequence and structural information with moderate accuracy. Here we develop a machine learning protocol for the prediction of DNA-BPs where the classifier is Support Vector Machines (SVMs). Information used for classification is derived from characteristics that include surface and overall composition, overall charge and positive potential patches on the protein surface. In total 121 DNA-BPs and 238 non-binding proteins are used to build and evaluate the protocol. In self-consistency, accuracy value of 100% has been achieved. For cross-validation (CV) optimization over entire dataset, we report an accuracy of 90%. Using leave 1-pair holdout evaluation, the accuracy of 86.3% has been achieved. When we restrict the dataset to less than 20% sequence identity amongst the proteins, the holdout accuracy is achieved at 85.8%. Furthermore, seven DNA-BPs with unbounded structures are all correctly predicted. The current performances are better than results published previously. The higher accuracy value achieved here originates from two factors: the ability of the SVM to handle features that demonstrate a wide range of discriminatory power and, a different definition of the positive patch. Since our protocol does not lean on sequence or structural homology, it can be used to identify or predict proteins with DNA-binding function(s) regardless of their homology to the known ones.