Function Prediction Methods Research Articles

CATH FunFHMMer web server: protein functional annotations using functional family assignments

The widening function annotation gap in protein databases and the increasing number and diversity of the proteins being sequenced presents new challenges to protein function prediction methods. Multidomain proteins complicate the protein sequence–structure–function relationship further as new combinations of domains can expand the functional repertoire, creating new proteins and functions. Here, we present the FunFHMMer web server, which provides Gene Ontology (GO) annotations for query protein sequences based on the functional classification of the domain-based CATH-Gene3D resource. Our server also provides valuable information for the prediction of functional sites. The predictive power of FunFHMMer has been validated on a set of 95 proteins where FunFHMMer performs better than BLAST, Pfam and CDD. Recent validation by an independent international competition ranks FunFHMMer as one of the top function prediction methods in predicting GO annotations for both the Biological Process and Molecular Function Ontology. The FunFHMMer web server is available at http://www.cathdb.info/search/by_funfhmmer.

VR-BFDT: A variance reduction based binary fuzzy decision tree induction method for protein function prediction

In protein function prediction (PFP) problem, the goal is to predict function of numerous well-sequenced known proteins whose function is not still known precisely. PFP is one of the special and complex problems in machine learning domain in which a protein (regarded as instance) may have more than one function simultaneously. Furthermore, the functions (regarded as classes) are dependent and also are organized in a hierarchical structure in the form of a tree or directed acyclic graph. One of the common learning methods proposed for solving this problem is decision trees in which, by partitioning data into sharp boundaries sets, small changes in the attribute values of a new instance may cause incorrect change in predicted label of the instance and finally misclassification. In this paper, a Variance Reduction based Binary Fuzzy Decision Tree (VR-BFDT) algorithm is proposed to predict functions of the proteins. This algorithm just fuzzifies the decision boundaries instead of converting the numeric attributes into fuzzy linguistic terms. It has the ability of assigning multiple functions to each protein simultaneously and preserves the hierarchy consistency between functional classes. It uses the label variance reduction as splitting criterion to select the best “attribute-value” at each node of the decision tree. The experimental results show that the overall performance of the proposed algorithm is promising.

IntFOLD: an integrated server for modelling protein structures and functions from amino acid sequences.

IntFOLD is an independent web server that integrates our leading methods for structure and function prediction. The server provides a simple unified interface that aims to make complex protein modelling data more accessible to life scientists. The server web interface is designed to be intuitive and integrates a complex set of quantitative data, so that 3D modelling results can be viewed on a single page and interpreted by non-expert modellers at a glance. The only required input to the server is an amino acid sequence for the target protein. Here we describe major performance and user interface updates to the server, which comprises an integrated pipeline of methods for: tertiary structure prediction, global and local 3D model quality assessment, disorder prediction, structural domain prediction, function prediction and modelling of protein-ligand interactions. The server has been independently validated during numerous CASP (Critical Assessment of Techniques for Protein Structure Prediction) experiments, as well as being continuously evaluated by the CAMEO (Continuous Automated Model Evaluation) project. The IntFOLD server is available at: http://www.reading.ac.uk/bioinf/IntFOLD/

Structural and functional modeling of protein in cumin (Cuminum cyminum)

Emergence of in-silico techniques have revolutionized the traditional methods of protein structure and function prediction of proteins. These computational tools encouraged the development of protein models in cumin (Cuminum cyminum L.) and have opened up new vistas in the area of research. In the present study, a high quality 3D structure and function of cumin RZ 19 protein have been predicted for the hypothetical amino acid sequence which showed homology with the protein domain of humans and E. coli illustrating that the database available on Apiaceae family is very low. The estimated molecular weight of identified cumin RZ 19 protein was 55028.6 and was predicted as an acidic protein with pI 5.14. Its functions are broadly classified into 2 categories: 1) Biological process: oxidationreduction and ion transport with a probability of 98.9% and 97.8% respectively and 2) Molecular function: structural constituent of ribosome with 99.0% and oxidoreductase activity with 98.7% probability. We have also identified a channel in cumin transmembrane protein, through which a ligand (an ion or a small molecule) might pass. The present finding may be a valuable addition to the proteomic information available on cumin. Further validation can be performed using wet lab experiments.

Probabilistic solution of floodplain inundation equation

Uncertainty in bed roughness is a dominant factor in providing a sufficiently accurate simulation of floodplain flows. This study describes a method to compute the transition probability density distribution of time-varying water elevations where the evolutionary process is based on a conventional one-dimensional storage cell model with governing stochastic differential equation. By including the random inputs (or noise terms) of bed roughness and initial water depth, time-dependent and spatially varying probability density function of the water surface leads to a Fokker–Planck equation. The model’s performance is evaluated by applying it to shallow water flow with a horizontal bed. Sensitivity of model predictions to variations in the bed friction parameters is shown. By comparing the result of the proposed method with that of conventional Monte Carlo simulation, the advantage of the former as a method for density function prediction is confirmed.

Hypergraph and Protein Function Prediction with Gene Expression Data

Most network-based protein (or gene) function prediction methods are based on the assumption that the labels of two adjacent proteins in the network are likely to be the same. However, assuming the pairwise relationship between proteins or genes is not complete. The information a group of genes that show very similar patterns of expression and tend to have similar functions (i.e. the functional modules) is missed. The natural way overcoming the information loss of the above assumption is to represent the gene expression data as the hypergraph. Thus, in this paper, the three un-normalized, random walk, and symmetric normalized hypergraph Laplacian based semi- supervised learning methods applied to hypergraph constructed from the gene expression data in order to predict the functions of yeast proteins are introduced. Experiment results show that the average accuracy performance measures of these three hypergraph Laplacian based semi-supervised learning methods are the same. However, their average accuracy performance measures of these three methods are much greater than the average accuracy performance measures of un-normalized graph Laplacian based semi-supervised learning method (i.e. the baseline method of this paper) applied to gene co-expression network created from the gene expression data. 

SHOPIN: Semantic Homogeneity Optimization in Protein Interaction Networks.

Protein interaction networks (PINs) are argued to be the richest source of hidden knowledge of the intrinsic physical and/or functional meanings of the involved proteins. We propose a novel method for computational protein function prediction based on semantic homogeneity optimization in PIN (SHOPIN). The SHOPIN method creates graph representations of the PIN augmented by inclusion of the semantics of the proteins and their interacting contexts. Network wide semantic relationships, modeled using random walks, are used to map the augmented PIN graphs in a new semantic metric space. The method produces a hierarchical partitioning of the PIN optimal in terms of semantic homogeneity by iterative optimization of the ratio of between clusters dissimilarities and within clusters similarities in the new semantic metric space. Function prediction is done using cluster wide-hierarchy high function enrichment. Results validate the rationale of the SHOPIN method placing it right next to state-of-the-art approaches performance wise.

PFP/ESG: automated protein function prediction servers enhanced with Gene Ontology visualization tool.

Protein function prediction (PFP) is an automated function prediction method that predicts Gene Ontology (GO) annotations for a protein sequence using distantly related sequences and contextual associations of GO terms. Extended similarity group (ESG) is another GO prediction algorithm that makes predictions based on iterative sequence database searches. Here, we provide interactive web servers for the PFP and ESG algorithms that are equipped with an effective visualization of the GO predictions in a hierarchical topology. PFP/ESG servers are freely available at http://kiharalab.org/web/pfp.php and http://kiharalab.org/web/esg.php, or access both at http://kiharalab.org/pfp_esg.php. Supplementary data are available at Bioinformatics online.

Study on the economic and environmental benefits of different EV powertrain topologies

• EV powertrain topologies can be realized by flexible distributed electric motors. • Model-based design optimization method is used to analyze and evaluate the EV. • The wheel-hub drive with reducer possesses the most competitive advantages. • US06 and NYCC result in the deterioration of EVs’ cost and GHG emissions. Numerous feasible schemes of powertrain topology can be designed for the electric vehicles (EVs) based on the distributed configurations of the electric motors. In this study, the effects of different EV powertrain topologies on the energy efficiency, vehicle ownership cost and lifecycle greenhouse gas (GHG) emissions of EVs are investigated. Energy-based vehicle simulation model including the regenerative braking function and battery degradation prediction method is established firstly. An optimization scheme combining the energy-based vehicle simulation model is conducted to minimize the electric energy consumption under various scenarios and driving conditions. Then the vehicle ownership cost and lifecycle GHG emissions of EVs are evaluated based on Chinese EV market and electricity grid. The sensitivity analyses of EV powertrain topology are implemented based on the different vehicle weights, CO 2 intensities of electricity and all-electric ranges. Results show that EVs using the powertrain of wheel-hub drive with the gear reducer have lower energy consumption. Furthermore, the driving cycles with more aggressive acceleration/deceleration and frequently stop-and-go conditions can increase both the vehicle ownership cost and lifecycle GHG emissions simultaneously. Chinese city traffic conditions will help EVs to obtain more benefits in respect of the economy and environment.

Negative example selection for protein function prediction: the NoGO database.

Negative examples – genes that are known not to carry out a given protein function – are rarely recorded in genome and proteome annotation databases, such as the Gene Ontology database. Negative examples are required, however, for several of the most powerful machine learning methods for integrative protein function prediction. Most protein function prediction efforts have relied on a variety of heuristics for the choice of negative examples. Determining the accuracy of methods for negative example prediction is itself a non-trivial task, given that the Open World Assumption as applied to gene annotations rules out many traditional validation metrics. We present a rigorous comparison of these heuristics, utilizing a temporal holdout, and a novel evaluation strategy for negative examples. We add to this comparison several algorithms adapted from Positive-Unlabeled learning scenarios in text-classification, which are the current state of the art methods for generating negative examples in low-density annotation contexts. Lastly, we present two novel algorithms of our own construction, one based on empirical conditional probability, and the other using topic modeling applied to genes and annotations. We demonstrate that our algorithms achieve significantly fewer incorrect negative example predictions than the current state of the art, using multiple benchmarks covering multiple organisms. Our methods may be applied to generate negative examples for any type of method that deals with protein function, and to this end we provide a database of negative examples in several well-studied organisms, for general use (The NoGO database, available at: bonneaulab.bio.nyu.edu/nogo.html).

Mutant p53 accumulation in human breast cancer is not an intrinsic property or dependent on structural or functional disruption but is regulated by exogenous stress and receptor status

Many human cancers contain missense TP53 mutations that result in p53 protein accumulation. Although generally considered as a single class of mutations that abrogate wild-type function, individual TP53 mutations may have specific properties and prognostic effects. Tumours that contain missense TP53 mutations show variable p53 stabilization patterns, which may reflect the specific mutation and/or aspects of tumour biology. We used immunohistochemistry on cell lines and human breast cancers with known TP53 missense mutations and assessed the effects of each mutation with four structure-function prediction methods. Cell lines with missense TP53 mutations show variable percentages of cells with p53 stabilization under normal growth conditions, ranging from approximately 50% to almost 100%. Stabilization is not related to structural or functional disruption, but agents that stabilize wild-type p53 increase the percentages of cells showing missense mutant p53 accumulation in cell lines with heterogeneous stabilization. The same heterogeneity of p53 stabilization occurs in primary breast cancers, independent of the effect of the mutation on structural properties or functional disruption. Heterogeneous accumulation is more common in steroid receptor-positive or HER2-positive breast cancers and cell lines than in triple-negative samples. Immunohistochemcal staining patterns associate with Mdm2 levels, proliferation, grade and overall survival, whilst the type of mutation reflects downstream target activity. Inhibiting Mdm2 activity increases the extent of p53 stabilization in some, but not all, breast cancer cell lines. The data indicate that missense mutant p53 stabilization is a complex and variable process in human breast cancers that associates with disease characteristics but is unrelated to structural or functional properties. That agents which stabilize wild-type p53 also stabilize mutant p53 has implications for patients with heterogeneous mutant p53 accumulation, where therapy may activate mutant p53 oncogenic function.

Neural network and SVM classifiers accurately predict lipid binding proteins, irrespective of sequence homology

Due to the central roles of lipid binding proteins (LBPs) in many biological processes, sequence based identification of LBPs is of great interest. The major challenge is that LBPs are diverse in sequence, structure, and function which results in low accuracy of sequence homology based methods. Therefore, there is a need for developing alternative functional prediction methods irrespective of sequence similarity. To identify LBPs from non-LBPs, the performances of support vector machine (SVM) and neural network were compared in this study. Comprehensive protein features and various techniques were employed to create datasets. Five-fold cross-validation (CV) and independent evaluation (IE) tests were used to assess the validity of the two methods. The results indicated that SVM outperforms neural network. SVM achieved 89.28% (CV) and 89.55% (IE) overall accuracy in identification of LBPs from non-LBPs and 92.06% (CV) and 92.90% (IE) (in average) for classification of different LBPs classes. Increasing the number and the range of extracted protein features as well as optimization of the SVM parameters significantly increased the efficiency of LBPs class prediction in comparison to the only previous report in this field. Altogether, the results showed that the SVM algorithm can be run on broad, computationally calculated protein features and offers a promising tool in detection of LBPs classes. The proposed approach has the potential to integrate and improve the common sequence alignment based methods.

Hierarchical ensemble methods for protein function prediction.

Protein function prediction is a complex multiclass multilabel classification problem, characterized by multiple issues such as the incompleteness of the available annotations, the integration of multiple sources of high dimensional biomolecular data, the unbalance of several functional classes, and the difficulty of univocally determining negative examples. Moreover, the hierarchical relationships between functional classes that characterize both the Gene Ontology and FunCat taxonomies motivate the development of hierarchy-aware prediction methods that showed significantly better performances than hierarchical-unaware “flat” prediction methods. In this paper, we provide a comprehensive review of hierarchical methods for protein function prediction based on ensembles of learning machines. According to this general approach, a separate learning machine is trained to learn a specific functional term and then the resulting predictions are assembled in a “consensus” ensemble decision, taking into account the hierarchical relationships between classes. The main hierarchical ensemble methods proposed in the literature are discussed in the context of existing computational methods for protein function prediction, highlighting their characteristics, advantages, and limitations. Open problems of this exciting research area of computational biology are finally considered, outlining novel perspectives for future research.

Protein Function Prediction with Incomplete Annotations.

Automated protein function prediction is one of the grand challenges in computational biology. Multi-label learning is widely used to predict functions of proteins. Most of multi-label learning methods make prediction for unlabeled proteins under the assumption that the labeled proteins are completely annotated, i.e., without any missing functions. However, in practice, we may have a subset of the ground-truth functions for a protein, and whether the protein has other functions is unknown. To predict protein functions with incomplete annotations, we propose a Protein Function Prediction method with Weak-label Learning (ProWL) and its variant ProWL-IF. Both ProWL and ProWL-IF can replenish the missing functions of proteins. In addition, ProWL-IF makes use of the knowledge that a protein cannot have certain functions, which can further boost the performance of protein function prediction. Our experimental results on protein-protein interaction networks and gene expression benchmarks validate the effectiveness of both ProWL and ProWL-IF.

Protein Function Prediction Using Semantic Driven K-Medoids Clustering Algorithm

The proposed protein function prediction methods are mostly based on sequence or structure protein similarity and do not take into account the similarity extracted from protein knowledge databases such as Gene Ontology. Many studies have shown that identification of protein complexes or functional modules can be effectively done by clustering protein interaction network (PIN). A significant number of proteins in such PIN remain uncharacterized and predicting their function remains a major challenge in system biology. In this paper we present a semantic driven clustering approach for protein function prediction by using both similarity metrics and the whole network topology of a PIN. We apply k-medoids clustering combined with several similarity metrics as a weight factor in the distance-clustering matrix. Protein functions are assigned based on cluster information. Results reveal improvement over standard non-semantic similarity metric.

Quantitative prediction of the effect of genetic variation using hidden Markov models

BackgroundWith the development of sequencing technologies, more and more sequence variants are available for investigation. Different classes of variants in the human genome have been identified, including single nucleotide substitutions, insertion and deletion, and large structural variations such as duplications and deletions. Insertion and deletion (indel) variants comprise a major proportion of human genetic variation. However, little is known about their effects on humans. The absence of understanding is largely due to the lack of both biological data and computational resources.ResultsThis paper presents a new indel functional prediction method HMMvar based on HMM profiles, which capture the conservation information in sequences. The results demonstrate that a scoring strategy based on HMM profiles can achieve good performance in identifying deleterious or neutral variants for different data sets, and can predict the protein functional effects of both single and multiple mutations.ConclusionsThis paper proposed a quantitative prediction method, HMMvar, to predict the effect of genetic variation using hidden Markov models. The HMM based pipeline program implementing the method HMMvar is freely available at https://bioinformatics.cs.vt.edu/zhanglab/hmm.

Prediction of Detailed Enzyme Functions and Identification of Specificity Determining Residues by Random Forests

Determining enzyme functions is essential for a thorough understanding of cellular processes. Although many prediction methods have been developed, it remains a significant challenge to predict enzyme functions at the fourth-digit level of the Enzyme Commission numbers. Functional specificity of enzymes often changes drastically by mutations of a small number of residues and therefore, information about these critical residues can potentially help discriminate detailed functions. However, because these residues must be identified by mutagenesis experiments, the available information is limited, and the lack of experimentally verified specificity determining residues (SDRs) has hindered the development of detailed function prediction methods and computational identification of SDRs. Here we present a novel method for predicting enzyme functions by random forests, EFPrf, along with a set of putative SDRs, the random forests derived SDRs (rf-SDRs). EFPrf consists of a set of binary predictors for enzymes in each CATH superfamily and the rf-SDRs are the residue positions corresponding to the most highly contributing attributes obtained from each predictor. EFPrf showed a precision of 0.98 and a recall of 0.89 in a cross-validated benchmark assessment. The rf-SDRs included many residues, whose importance for specificity had been validated experimentally. The analysis of the rf-SDRs revealed both a general tendency that functionally diverged superfamilies tend to include more active site residues in their rf-SDRs than in less diverged superfamilies, and superfamily-specific conservation patterns of each functional residue. EFPrf and the rf-SDRs will be an effective tool for annotating enzyme functions and for understanding how enzyme functions have diverged within each superfamily.

A novel method for functional annotation prediction based on combination of classification methods.

Automated protein function prediction defines the designation of functions of unknown protein functions by using computational methods. This technique is useful to automatically assign gene functional annotations for undefined sequences in next generation genome analysis (NGS). NGS is a popular research method since high-throughput technologies such as DNA sequencing and microarrays have created large sets of genes. These huge sequences have greatly increased the need for analysis. Previous research has been based on the similarities of sequences as this is strongly related to the functional homology. However, this study aimed to designate protein functions by automatically predicting the function of the genome by utilizing InterPro (IPR), which can represent the properties of the protein family and groups of the protein function. Moreover, we used gene ontology (GO), which is the controlled vocabulary used to comprehensively describe the protein function. To define the relationship between IPR and GO terms, three pattern recognition techniques have been employed under different conditions, such as feature selection and weighted value, instead of a binary one.

Profiling the orphan enzymes

The emergence of Next Generation Sequencing generates an incredible amount of sequence and great potential for new enzyme discovery. Despite this huge amount of data and the profusion of bioinformatic methods for function prediction, a large part of known enzyme activities is still lacking an associated protein sequence. These particular activities are called “orphan enzymes”. The present review proposes an update of previous surveys on orphan enzymes by mining the current content of public databases. While the percentage of orphan enzyme activities has decreased from 38% to 22% in ten years, there are still more than 1,000 orphans among the 5,000 entries of the Enzyme Commission (EC) classification. Taking into account all the reactions present in metabolic databases, this proportion dramatically increases to reach nearly 50% of orphans and many of them are not associated to a known pathway. We extended our survey to “local orphan enzymes” that are activities which have no representative sequence in a given clade, but have at least one in organisms belonging to other clades. We observe an important bias in Archaea and find that in general more than 30% of the EC activities have incomplete sequence information in at least one superkingdom. To estimate if candidate proteins for local orphans could be retrieved by homology search, we applied a simple strategy based on the PRIAM software and noticed that candidates may be proposed for an important fraction of local orphan enzymes. Finally, by studying relation between protein domains and catalyzed activities, it appears that newly discovered enzymes are mostly associated with already known enzyme domains. Thus, the exploration of the promiscuity and the multifunctional aspect of known enzyme families may solve part of the orphan enzyme issue. We conclude this review with a presentation of recent initiatives in finding proteins for orphan enzymes and in extending the enzyme world by the discovery of new activities.ReviewersThis article was reviewed by Michael Galperin, Daniel Haft and Daniel Kahn.

FunPred-1: protein function prediction from a protein interaction network using neighborhood analysis.

Proteins are responsible for all biological activities in living organisms. Thanks to genome sequencing projects, large amounts of DNA and protein sequence data are now available, but the biological functions of many proteins are still not annotated in most cases. The unknown function of such non-annotated proteins may be inferred or deduced from their neighbors in a protein interaction network. In this paper, we propose two new methods to predict protein functions based on network neighborhood properties. FunPred 1.1 uses a combination of three simple-yet-effective scoring techniques: the neighborhood ratio, the protein path connectivity and the relative functional similarity. FunPred 1.2 applies a heuristic approach using the edge clustering coefficient to reduce the search space by identifying densely connected neighborhood regions. The overall accuracy achieved in FunPred 1.2 over 8 functional groups involving hetero-interactions in 650 yeast proteins is around 87%, which is higher than the accuracy with FunPred 1.1. It is also higher than the accuracy of many of the state-of-the-art protein function prediction methods described in the literature. The test datasets and the complete source code of the developed software are now freely available at http://code.google.com/p/cmaterbioinfo/.