Accurate Function Prediction Research Articles

EnzyMine: a comprehensive database for enzyme function annotation with enzymatic reaction chemical feature.

Addition of chemical structural information in enzymatic reactions has proven to be significant for accurate enzyme function prediction. However, such chemical data lack systematic feature mining and hardly exist in enzyme-related databases. Therefore, global mining of enzymatic reactions will offer a unique landscape for researchers to understand the basic functional mechanisms of natural bioprocesses and facilitate enzyme function annotation. Here, we established a new knowledge base called EnzyMine, through which we propose to elucidate enzymatic reaction features and then link them with sequence and structural annotations. EnzyMine represents an advanced database that extends enzyme knowledge by incorporating reaction chemical feature strategies, strengthening the connectivity between enzyme and metabolic reactions. Therefore, it has the potential to reveal many new metabolic pathways involved with given enzymes, as well as expand enzyme function annotation. Database URL: http://www.rxnfinder.org/enzymine/.

DeepGOPlus: improved protein function prediction from sequence.

Protein function prediction is one of the major tasks of bioinformatics that can help in wide range of biological problems such as understanding disease mechanisms or finding drug targets. Many methods are available for predicting protein functions from sequence based features, protein-protein interaction networks, protein structure or literature. However, other than sequence, most of the features are difficult to obtain or not available for many proteins thereby limiting their scope. Furthermore, the performance of sequence-based function prediction methods is often lower than methods that incorporate multiple features and predicting protein functions may require a lot of time. We developed a novel method for predicting protein functions from sequence alone which combines deep convolutional neural network (CNN) model with sequence similarity based predictions. Our CNN model scans the sequence for motifs which are predictive for protein functions and combines this with functions of similar proteins (if available). We evaluate the performance of DeepGOPlus using the CAFA3 evaluation measures and achieve an Fmax of 0.390, 0.557 and 0.614 for BPO, MFO and CCO evaluations, respectively. These results would have made DeepGOPlus one of the three best predictors in CCO and the second best performing method in the BPO and MFO evaluations. We also compare DeepGOPlus with state-of-the-art methods such as DeepText2GO and GOLabeler on another dataset. DeepGOPlus can annotate around 40 protein sequences per second on common hardware, thereby making fast and accurate function predictions available for a wide range of proteins. http://deepgoplus.bio2vec.net/ . Supplementary data are available at Bioinformatics online.

Structural basis of the nonribosomal codes for nonproteinogenic amino acid selective adenylation enzymes in the biosynthesis of natural products.

Nonproteinogenic amino acids are the unique building blocks of nonribosomal peptides (NRPs) and hybrid nonribosomal peptide-polyketides (NRP-PKs) and contribute to their diversity of chemical structures and biological activities. In the biosynthesis of NRPs and NRP-PKs, adenylation enzymes select and activate an amino acid substrate as an aminoacyl adenylate, which reacts with the thiol of the holo form of the carrier protein to afford an aminoacyl thioester as the electrophile for the condensation reaction. Therefore, the substrate specificity of adenylation enzymes is a key determinant of the structure of NRPs and NRP-PKs. Here, we focus on nonproteinogenic amino acid selective adenylation enzymes, because understanding their unique selection mechanisms will lead to accurate functional predictions and protein engineering toward the rational biosynthesis of designed molecules containing amino acids. Based on recent progress in the structural analysis of adenylation enzymes, we discuss the nonribosomal codes of nonproteinogenic amino acid selective adenylation enzymes.

A semi-supervised approach for predicting cell-type specific functional consequences of non-coding variation using MPRAs

Predicting the functional consequences of genetic variants in non-coding regions is a challenging problem. We propose here a semi-supervised approach, GenoNet, to jointly utilize experimentally confirmed regulatory variants (labeled variants), millions of unlabeled variants genome-wide, and more than a thousand cell/tissue type specific epigenetic annotations to predict functional consequences of non-coding variants. Through the application to several experimental datasets, we demonstrate that the proposed method significantly improves prediction accuracy compared to existing functional prediction methods at the tissue/cell type level, but especially so at the organism level. Importantly, we illustrate how the GenoNet scores can help in fine-mapping at GWAS loci, and in the discovery of disease associated genes in sequencing studies. As more comprehensive lists of experimentally validated variants become available over the next few years, semi-supervised methods like GenoNet can be used to provide increasingly accurate functional predictions for variants genome-wide and across a variety of cell/tissue types.

Optimizing special quasirandom structure (SQS) models for accurate functional property prediction in disordered 2D alloys

2D materials such as MXenes have garnered attention in a wide field of applications ranging from energy to environment to medical. Properties of 2D materials can be tailored via alloying and in some cases, solid-solutions (disordered alloys) are formed. To predict the disordered alloy properties via first-principles, the model structure needs to imitate the random arrangements of alloyants and yet remains computationally tractable. Using density functional theory and the cluster expansion method, we investigate the accuracy of using of special quasirandom structures (SQSs) for predicting disordered 2D alloy properties, evaluating the effect of SQS supercell size on the prediction quality of formation energies, elastic properties, and structural parameters. We illustrate the findings with 5 different disordered binary MXene alloy systems (where M = Ti and M′ = Zr, Hf, V, Nb, or Ta), demonstrating that SQSs around 6–8 times the primitive cell (N = 6–8) are sufficient to attain convergence in the property predictions versus supercell size. For formation energies, SQSs with N > 4 are found to reproduce the formation energies of the fully disordered phase within ~2.5 meV. For the simulation of the experimentally-synthesized TiNbCO2, we find convergence in structural parameters and elastic tensors at N ~ 6. We traced the convergence of the predictions to the convergence in the band structure-related properties via analysis of the electronic densities-of-states and the projected crystal overlap Hamilton population. Our findings suggest that modest sized SQSs would reproduce the properties of disordered MXene alloys. The results should help guide the investigations of structure-property relationships in other disordered 2D materials as well.

DeepFam: deep learning based alignment-free method for protein family modeling and prediction.

MotivationA large number of newly sequenced proteins are generated by the next-generation sequencing technologies and the biochemical function assignment of the proteins is an important task. However, biological experiments are too expensive to characterize such a large number of protein sequences, thus protein function prediction is primarily done by computational modeling methods, such as profile Hidden Markov Model (pHMM) and k-mer based methods. Nevertheless, existing methods have some limitations; k-mer based methods are not accurate enough to assign protein functions and pHMM is not fast enough to handle large number of protein sequences from numerous genome projects. Therefore, a more accurate and faster protein function prediction method is needed.ResultsIn this paper, we introduce DeepFam, an alignment-free method that can extract functional information directly from sequences without the need of multiple sequence alignments. In extensive experiments using the Clusters of Orthologous Groups (COGs) and G protein-coupled receptor (GPCR) dataset, DeepFam achieved better performance in terms of accuracy and runtime for predicting functions of proteins compared to the state-of-the-art methods, both alignment-free and alignment-based methods. Additionally, we showed that DeepFam has a power of capturing conserved regions to model protein families. In fact, DeepFam was able to detect conserved regions documented in the Prosite database while predicting functions of proteins. Our deep learning method will be useful in characterizing functions of the ever increasing protein sequences.Availability and implementationCodes are available at https://bhi-kimlab.github.io/DeepFam.

Accurate density functional prediction of molecular electron affinity with the scaling corrected Kohn–Sham frontier orbital energies

ABSTRACTElectron affinity (EA) is the energy released when an additional electron is attached to an atom or a molecule. EA is a fundamental thermochemical property, and it is closely pertinent to other important properties such as electronegativity and hardness. However, accurate prediction of EA is difficult with density functional theory methods. The somewhat large error of the calculated EAs originates mainly from the intrinsic delocalisation error associated with the approximate exchange--correlation functional. In this work, we employ a previously developed non-empirical global scaling correction approach, which explicitly imposes the Perdew–Parr–Levy–Balduz condition to the approximate functional, and achieve a substantially improved accuracy for the calculated EAs. In our approach, the EA is given by the scaling corrected Kohn–Sham lowest unoccupied molecular orbital energy of the neutral molecule, without the need to carry out the self-consistent-field calculation for the anion.

Evaluating Functional Annotations of Enzymes Using the Gene Ontology.

The Gene Ontology (GO) (Ashburner et al., Nat Genet 25(1):25–29, 2000) is a powerful tool in the informatics arsenal of methods for evaluating annotations in a protein dataset. From identifying the nearest well annotated homologue of a protein of interest to predicting where misannotation has occurred to knowing how confident you can be in the annotations assigned to those proteins is critical. In this chapter we explore what makes an enzyme unique and how we can use GO to infer aspects of protein function based on sequence similarity. These can range from identification of misannotation or other errors in a predicted function to accurate function prediction for an enzyme of entirely unknown function. Although GO annotation applies to any gene products, we focus here a describing our approach for hierarchical classification of enzymes in the Structure-Function Linkage Database (SFLD) (Akiva et al., Nucleic Acids Res 42(Database issue):D521–530, 2014) as a guide for informed utilisation of annotation transfer based on GO terms.

SIFTER search: a web server for accurate phylogeny-based protein function prediction.

We are awash in proteins discovered through high-throughput sequencing projects. As only a minuscule fraction of these have been experimentally characterized, computational methods are widely used for automated annotation. Here, we introduce a user-friendly web interface for accurate protein function prediction using the SIFTER algorithm. SIFTER is a state-of-the-art sequence-based gene molecular function prediction algorithm that uses a statistical model of function evolution to incorporate annotations throughout the phylogenetic tree. Due to the resources needed by the SIFTER algorithm, running SIFTER locally is not trivial for most users, especially for large-scale problems. The SIFTER web server thus provides access to precomputed predictions on 16 863 537 proteins from 232 403 species. Users can explore SIFTER predictions with queries for proteins, species, functions, and homologs of sequences not in the precomputed prediction set. The SIFTER web server is accessible at http://sifter.berkeley.edu/ and the source code can be downloaded.

Expanding our understanding of sequence-function relationships of type II polyketide biosynthetic gene clusters: bioinformatics-guided identification of Frankiamicin A from Frankia sp. EAN1pec.

A large and rapidly increasing number of unstudied “orphan” natural product biosynthetic gene clusters are being uncovered in sequenced microbial genomes. An important goal of modern natural products research is to be able to accurately predict natural product structures and biosynthetic pathways from these gene cluster sequences. This requires both development of bioinformatic methods for global analysis of these gene clusters and experimental characterization of select products produced by gene clusters with divergent sequence characteristics. Here, we conduct global bioinformatic analysis of all available type II polyketide gene cluster sequences and identify a conserved set of gene clusters with unique ketosynthase α/β sequence characteristics in the genomes of Frankia species, a group of Actinobacteria with underexploited natural product biosynthetic potential. Through LC-MS profiling of extracts from several Frankia species grown under various conditions, we identified Frankia sp. EAN1pec as producing a compound with spectral characteristics consistent with the type II polyketide produced by this gene cluster. We isolated the compound, a pentangular polyketide which we named frankiamicin A, and elucidated its structure by NMR and labeled precursor feeding. We also propose biosynthetic and regulatory pathways for frankiamicin A based on comparative genomic analysis and literature precedent, and conduct bioactivity assays of the compound. Our findings provide new information linking this set of Frankia gene clusters with the compound they produce, and our approach has implications for accurate functional prediction of the many other type II polyketide clusters present in bacterial genomes.

Automatic classification of highly related Malate Dehydrogenase and L-Lactate Dehydrogenase based on 3D-pattern of active sites

Accurate protein function prediction is an important subject in bioinformatics, especially wheresequentially and structurally similar proteins have different functions. Malate dehydrogenaseand L-lactate dehydrogenase are two evolutionary related enzymes, which exist in a widevariety of organisms. These enzymes are sequentially and structurally similar and sharecommon active site residues, spatial patterns and molecular mechanisms. Here, we studyvarious features of the active site cavity of 229 PDB chain entries and try to classify themautomatically by various classifiers including the support vector machine, k nearest neighbourand random forest methods. The results show that the support vector machine yields the highestpredictive performance among mentioned classifiers. Despite very close and conserved patternsamong Malate dehydrogenases and L-lactate dehydrogenases, the SVM predicts the functionefficiently and achieves 0.973 Matthew’s correlation coefficient and 0.987 F-score. The sameapproach can be used in other enzyme families for automatic discrimination betweenhomologous enzymes with common active site elements, however, acting on differentsubstrates.

A New Protein Structure Representation for Efficient Protein Function Prediction

One of the challenging problems in bioinformatics is the prediction of protein function. Protein function is the main key that can be used to classify different proteins. Protein function can be inferred experimentally with very small throughput or computationally with very high throughput. Computational methods are sequence based or structure based. Structure-based methods produce more accurate protein function prediction. In this article, we propose a new protein structure representation for efficient protein function prediction. The representation is based on three-dimensional patterns of protein residues. In the analysis, we used protein function based on enzyme activity through six mechanistically diverse enzyme superfamilies: amidohydrolase, crotonase, haloacid dehalogenase, isoprenoid synthase type I, and vicinal oxygen chelate. We applied three different classification methods, naïve Bayes, k-nearest neighbors, and random forest, to predict the enzyme superfamily of a given protein. The prediction accuracy using the proposed representation outperforms a recently introduced representation method that is based only on the distance patterns. The results show that the proposed representation achieved prediction accuracy up to 98%, with improvement of about 10% on average.

Hidden Information Revealed by Optimal Community Structure from a Protein-Complex Bipartite Network Improves Protein Function Prediction

The task of extracting the maximal amount of information from a biological network has drawn much attention from researchers, for example, predicting the function of a protein from a protein-protein interaction (PPI) network. It is well known that biological networks consist of modules/communities, a set of nodes that are more densely inter-connected among themselves than with the rest of the network. However, practical applications of utilizing the community information have been rather limited. For protein function prediction on a network, it has been shown that none of the existing community-based protein function prediction methods outperform a simple neighbor-based method. Recently, we have shown that proper utilization of a highly optimal modularity community structure for protein function prediction can outperform neighbor-assisted methods. In this study, we propose two function prediction approaches on bipartite networks that consider the community structure information as well as the neighbor information from the network: 1) a simple screening method and 2) a random forest based method. We demonstrate that our community-assisted methods outperform neighbor-assisted methods and the random forest method yields the best performance. In addition, we show that using the optimal community structure information is essential for more accurate function prediction for the protein-complex bipartite network of Saccharomyces cerevisiae. Community detection can be carried out either using a modified modularity for dealing with the original bipartite network or first projecting the network into a single-mode network (i.e., PPI network) and then applying community detection to the reduced network. We find that the projection leads to the loss of information in a significant way. Since our prediction methods rely only on the network topology, they can be applied to various fields where an efficient network-based analysis is required.

Functional Knowledge Transfer for High-accuracy Prediction of Under-studied Biological Processes

A key challenge in genetics is identifying the functional roles of genes in pathways. Numerous functional genomics techniques (e.g. machine learning) that predict protein function have been developed to address this question. These methods generally build from existing annotations of genes to pathways and thus are often unable to identify additional genes participating in processes that are not already well studied. Many of these processes are well studied in some organism, but not necessarily in an investigator's organism of interest. Sequence-based search methods (e.g. BLAST) have been used to transfer such annotation information between organisms. We demonstrate that functional genomics can complement traditional sequence similarity to improve the transfer of gene annotations between organisms. Our method transfers annotations only when functionally appropriate as determined by genomic data and can be used with any prediction algorithm to combine transferred gene function knowledge with organism-specific high-throughput data to enable accurate function prediction.We show that diverse state-of-art machine learning algorithms leveraging functional knowledge transfer (FKT) dramatically improve their accuracy in predicting gene-pathway membership, particularly for processes with little experimental knowledge in an organism. We also show that our method compares favorably to annotation transfer by sequence similarity. Next, we deploy FKT with state-of-the-art SVM classifier to predict novel genes to 11,000 biological processes across six diverse organisms and expand the coverage of accurate function predictions to processes that are often ignored because of a dearth of annotated genes in an organism. Finally, we perform in vivo experimental investigation in Danio rerio and confirm the regulatory role of our top predicted novel gene, wnt5b, in leftward cell migration during heart development. FKT is immediately applicable to many bioinformatics techniques and will help biologists systematically integrate prior knowledge from diverse systems to direct targeted experiments in their organism of study.

IMP: a multi-species functional genomics portal for integration, visualization and prediction of protein functions and networks

Integrative multi-species prediction (IMP) is an interactive web server that enables molecular biologists to interpret experimental results and to generate hypotheses in the context of a large cross-organism compendium of functional predictions and networks. The system provides a framework for biologists to analyze their candidate gene sets in the context of functional networks, as they expand or focus these sets by mining functional relationships predicted from integrated high-throughput data. IMP integrates prior knowledge and data collections from multiple organisms in its analyses. Through flexible and interactive visualizations, researchers can compare functional contexts and interpret the behavior of their gene sets across organisms. Additionally, IMP identifies homologs with conserved functional roles for knowledge transfer, allowing for accurate function predictions even for biological processes that have very few experimental annotations in a given organism. IMP currently supports seven organisms (Homo sapiens, Mus musculus, Rattus novegicus, Drosophila melanogaster, Danio rerio, Caenorhabditis elegans and Saccharomyces cerevisiae), does not require any registration or installation and is freely available for use at http://imp.princeton.edu.

Highly accurate and high-resolution function prediction of RNA binding proteins by fold recognition and binding affinity prediction

A full understanding of the mechanism of post- transcriptional regulation requires more than simple two- state prediction (binding or not binding) for RNA binding proteins. Here we report a sequence-based technique dedicated for predicting complex structures of protein and RNA by combining fold recognition with binding affinity prediction. The method not only provides a highly accurate complex structure prediction (77% of residues are within 4°A RMSD from native in average for the independent test set) but also achieves the best performing two-state binding or non-binding prediction with an accuracy of 98%, precision of 84%, and Mathews correlation coefficient (MCC) of 0.62. Moreover, it predicts binding residues with an accuracy of 84%, precision of 66% and MCC value of 0.51. In addition, it has a success rate of 77% in predicting RNA binding types (mRNA, tRNA or rRNA). We further demonstrate that it makes more than 10% improvement either in precision or sensitivity than PSI- BLAST, HHPRED and our previously developed structure- based technique. This method expects to be useful for highly accurate genome-scale, high-resolution prediction of RNA-binding proteins and their complex structures. A web server (SPOT) is freely available for academic users at http://sparks.informatics.iupui.edu.

Adaptive multi-agent architecture for functional sequence motifs recognition

Accurate genome annotation or protein function prediction requires precise recognition of functional sequence motifs. Many computational motif prediction models have been proposed. Due to the complexity of the biological data, it may be desirable to apply an integrated approach that uses multiple models for analysis. In this article, we propose a novel multi-agent architecture for the general purpose of functional sequence motif recognition. The approach takes advantage of the synergy provided by multiple agents through the employment of different agents equipped with distinctive problem solving skills and promotes the collaborations among them through decision maker (DM) agents that work as classifier ensembles. A genetic algorithm-based fusion strategy is applied which offers evolutionary property to the DM agents. The consistency and robustness of the system are maintained by an evolvable agent that mediates the team of the ensemble agents. The combined effort of a recommendation system (Seer) and the self-learning mediator agent yields a successful identification of the most efficient agent deployment scheme at an early stage of the experimentation process, which has the potential of greatly reducing the computational cost of the system. Two concrete systems are constructed that aim at predicting two important sequence motifs-the translational initiation sites (TISs) and the core promoters. With the incorporation of three distinctive problem solver agents, the TIS predictor consistently outperforms most of the state-of-the-art approaches under investigation. Integrating three existing promoter predictors, our system is able to yield consistently good performance. The program (MotifMAS) and the datasets are available upon request.

Phylogenetic molecular function annotation

It is now easier to discover thousands of protein sequences in a new microbial genome than it is to biochemically characterize the specific activity of a single protein of unknown function. The molecular functions of protein sequences have typically been predicted using homology-based computational methods, which rely on the principle that homologous proteins share a similar function. However, some protein families include groups of proteins with different molecular functions. A phylogenetic approach for predicting molecular function (sometimes called "phylogenomics") is an effective means to predict protein molecular function. These methods incorporate functional evidence from all members of a family that have functional characterizations using the evolutionary history of the protein family to make robust predictions for the uncharacterized proteins. However, they are often difficult to apply on a genome-wide scale because of the time-consuming step of reconstructing the phylogenies of each protein to be annotated. Our automated approach for function annotation using phylogeny, the SIFTER (Statistical Inference of Function Through Evolutionary Relationships) methodology, uses a statistical graphical model to compute the probabilities of molecular functions for unannotated proteins. Our benchmark tests showed that SIFTER provides accurate functional predictions on various protein families, outperforming other available methods.

Towards structured output prediction of enzyme function.

In this paper we describe work in progress in developing kernel methods for enzyme function prediction. Our focus is in developing so called structured output prediction methods, where the enzymatic reaction is the combinatorial target object for prediction. We compared two structured output prediction methods, the Hierarchical Max-Margin Markov algorithm (HM3) and the Maximum Margin Regression algorithm (MMR) in hierarchical classification of enzyme function. As sequence features we use various string kernels and the GTG feature set derived from the global alignment trace graph of protein sequences. In our experiments, in predicting enzyme EC classification we obtain over 85% accuracy (predicting the four digit EC code) and over 91% microlabel F1 score (predicting individual EC digits). In predicting the Gold Standard enzyme families, we obtain over 79% accuracy (predicting family correctly) and over 89% microlabel F1 score (predicting superfamilies and families). In the latter case, structured output methods are significantly more accurate than nearest neighbor classifier. A polynomial kernel over the GTG feature set turned out to be a prerequisite for accurate function prediction. Combining GTG with string kernels boosted accuracy slightly in the case of EC class prediction. Structured output prediction with GTG features is shown to be computationally feasible and to have accuracy on par with state-of-the-art approaches in enzyme function prediction.

Discovering functional linkages and uncharacterized cellular pathways using phylogenetic profile comparisons: a comprehensive assessment

BackgroundA widely-used approach for discovering functional and physical interactions among proteins involves phylogenetic profile comparisons (PPCs). Here, proteins with similar profiles are inferred to be functionally related under the assumption that proteins involved in the same metabolic pathway or cellular system are likely to have been co-inherited during evolution.ResultsOur experimentation with E. coli and yeast proteins with 16 different carefully composed reference sets of genomes revealed that the phyletic patterns of proteins in prokaryotes alone could be adequate enough to make reasonably accurate functional linkage predictions. A slight improvement in performance is observed on adding few eukaryotes into the reference set, but a noticeable drop-off in performance is observed with increased number of eukaryotes. Inclusion of most parasitic, pathogenic or vertebrate genomes and multiple strains of the same species into the reference set do not necessarily contribute to an improved sensitivity or accuracy. Interestingly, we also found that evolutionary histories of individual pathways have a significant affect on the performance of the PPC approach with respect to a particular reference set. For example, to accurately predict functional links in carbohydrate or lipid metabolism, a reference set solely composed of prokaryotic (or bacterial) genomes performed among the best compared to one composed of genomes from all three super-kingdoms; this is in contrast to predicting functional links in translation for which a reference set composed of prokaryotic (or bacterial) genomes performed the worst. We also demonstrate that the widely used random null model to quantify the statistical significance of profile similarity is incomplete, which could result in an increased number of false-positives.ConclusionContrary to previous proposals, it is not merely the number of genomes but a careful selection of informative genomes in the reference set that influences the prediction accuracy of the PPC approach. We note that the predictive power of the PPC approach, especially in eukaryotes, is heavily influenced by the primary endosymbiosis and subsequent bacterial contributions. The over-representation of parasitic unicellular eukaryotes and vertebrates additionally make eukaryotes less useful in the reference sets. Reference sets composed of highly non-redundant set of genomes from all three super-kingdoms fare better with pathways showing considerable vertical inheritance and strong conservation (e.g. translation apparatus), while reference sets solely composed of prokaryotic genomes fare better for more variable pathways like carbohydrate metabolism. Differential performance of the PPC approach on various pathways, and a weak positive correlation between functional and profile similarities suggest that caution should be exercised while interpreting functional linkages inferred from genome-wide large-scale profile comparisons using a single reference set.