Automated Function Prediction Research Articles

NetGO 3.0: Protein Language Model Improves Large-scale Functional Annotations

As one of the state-of-the-art automated function prediction (AFP) methods, NetGO 2.0 integrates multi-source information to improve the performance. However, it mainly utilizes the proteins with experimentally supported functional annotations without leveraging valuable information from a vast number of unannotated proteins. Recently, protein language models have been proposed to learn informative representations [e.g., Evolutionary Scale Modeling (ESM)-1b embedding] from protein sequences based on self-supervision. Here, we represented each protein by ESM-1b and used logistic regression (LR) to train a new model, LR-ESM, for AFP. The experimental results showed that LR-ESM achieved comparable performance with the best-performing component of NetGO 2.0. Therefore, by incorporating LR-ESM into NetGO 2.0, we developed NetGO 3.0 to improve the performance of AFP extensively. NetGO 3.0 is freely accessible at https://dmiip.sjtu.edu.cn/ng3.0.

Gene function prediction in five model eukaryotes exclusively based on gene relative location through machine learning

The function of most genes is unknown. The best results in automated function prediction are obtained with machine learning-based methods that combine multiple data sources, typically sequence derived features, protein structure and interaction data. Even though there is ample evidence showing that a gene’s function is not independent of its location, the few available examples of gene function prediction based on gene location rely on sequence identity between genes of different organisms and are thus subjected to the limitations of the relationship between sequence and function. Here we predict thousands of gene functions in five model eukaryotes (Saccharomyces cerevisiae, Caenorhabditis elegans, Drosophila melanogaster, Mus musculus and Homo sapiens) using machine learning models exclusively trained with features derived from the location of genes in the genomes to which they belong. Our aim was not to obtain the best performing method to automated function prediction but to explore the extent to which a gene's location can predict its function in eukaryotes. We found that our models outperform BLAST when predicting terms from Biological Process and Cellular Component Ontologies, showing that, at least in some cases, gene location alone can be more useful than sequence to infer gene function.

PANNZER-A practical tool for protein function prediction.

The facility of next-generation sequencing has led to an explosion of gene catalogs for novel genomes, transcriptomes and metagenomes, which are functionally uncharacterized. Computational inference has emerged as a necessary substitute for first-hand experimental evidence. PANNZER (Protein ANNotation with Z-scoRE) is a high-throughput functional annotation web server that stands out among similar publically accessible web servers in supporting submission of up to 100,000 protein sequences at once and providing both Gene Ontology (GO) annotations and free text description predictions. Here, we demonstrate the use of PANNZER and discuss future plans and challenges. We present two case studies to illustrate problems related to data quality and method evaluation. Some commonly used evaluation metrics and evaluation datasets promote methods that favor unspecific and broad functional classes over more informative and specific classes. We argue that this can bias the development of automated function prediction methods. The PANNZER web server and source code are available at http://ekhidna2.biocenter.helsinki.fi/sanspanz/.

Multi-omics network-based functional annotation of unknown Arabidopsis genes.

Unraveling gene function is pivotal to understanding the signaling cascades that control plant development and stress responses. As experimental profiling is costly and labor intensive, there is a clear need for high-confidence computational annotation. In contrast to detailed gene-specific functional information, transcriptomics data are widely available for both model and crop species. Here, we describe a novel automated function prediction method, which leverages complementary information from multiple expression datasets by analyzing study-specific gene co-expression networks. First, we benchmarked the prediction performance on recently characterized Arabidopsis thaliana genes, and showed that our method outperforms state-of-the-art expression-based approaches. Next, we predicted biological process annotations for known (n = 15 790) and unknown (n = 11 865) genes in A. thaliana and validated our predictions using experimental protein-DNA and protein-protein interaction data (covering >220 000 interactions in total), obtaining a set of high-confidence functional annotations. Our method assigned at least one validated annotation to 5054 (42.6%) unknown genes, and at least one novel validated function to 3408 (53.0%) genes with computational annotations only. These omics-supported functional annotations shed light on a variety of developmental processes and molecular responses, such as flower and root development, defense responses to fungi and bacteria, and phytohormone signaling, and help fill the information gap on biological process annotations in Arabidopsis. An in-depth analysis of two context-specific networks, modeling seed development and response to water deprivation, shows how previously uncharacterized genes function within the respective networks. Moreover, our automated function prediction approach can be applied in future studies to facilitate gene discovery for crop improvement.

Structure-based protein function prediction using graph convolutional networks

The rapid increase in the number of proteins in sequence databases and the diversity of their functions challenge computational approaches for automated function prediction. Here, we introduce DeepFRI, a Graph Convolutional Network for predicting protein functions by leveraging sequence features extracted from a protein language model and protein structures. It outperforms current leading methods and sequence-based Convolutional Neural Networks and scales to the size of current sequence repositories. Augmenting the training set of experimental structures with homology models allows us to significantly expand the number of predictable functions. DeepFRI has significant de-noising capability, with only a minor drop in performance when experimental structures are replaced by protein models. Class activation mapping allows function predictions at an unprecedented resolution, allowing site-specific annotations at the residue-level in an automated manner. We show the utility and high performance of our method by annotating structures from the PDB and SWISS-MODEL, making several new confident function predictions. DeepFRI is available as a webserver at https://beta.deepfri.flatironinstitute.org/.

GOLabeler: improving sequence-based large-scale protein function prediction by learning to rank.

Gene Ontology (GO) has been widely used to annotate functions of proteins and understand their biological roles. Currently only <1% of >70 million proteins in UniProtKB have experimental GO annotations, implying the strong necessity of automated function prediction (AFP) of proteins, where AFP is a hard multilabel classification problem due to one protein with a diverse number of GO terms. Most of these proteins have only sequences as input information, indicating the importance of sequence-based AFP (SAFP: sequences are the only input). Furthermore, homology-based SAFP tools are competitive in AFP competitions, while they do not necessarily work well for so-called difficult proteins, which have <60% sequence identity to proteins with annotations already. Thus, the vital and challenging problem now is how to develop a method for SAFP, particularly for difficult proteins. The key of this method is to extract not only homology information but also diverse, deep-rooted information/evidence from sequence inputs and integrate them into a predictor in a both effective and efficient manner. We propose GOLabeler, which integrates five component classifiers, trained from different features, including GO term frequency, sequence alignment, amino acid trigram, domains and motifs, and biophysical properties, etc., in the framework of learning to rank (LTR), a paradigm of machine learning, especially powerful for multilabel classification. The empirical results obtained by examining GOLabeler extensively and thoroughly by using large-scale datasets revealed numerous favorable aspects of GOLabeler, including significant performance advantage over state-of-the-art AFP methods. http://datamining-iip.fudan.edu.cn/golabeler. Supplementary data are available at Bioinformatics online.

Large-scale automated function prediction of protein sequences and an experimental case study validation on PTEN transcript variants.

Recent advances in computing power and machine learning empower functional annotation of protein sequences and their transcript variations. Here, we present an automated prediction system UniGOPred, for GO annotations and a database of GO term predictions for proteomes of several organisms in UniProt Knowledgebase (UniProtKB). UniGOPred provides function predictions for 514 molecular function (MF), 2909 biological process (BP), and 438 cellular component (CC) GO terms for each protein sequence. UniGOPred covers nearly the whole functionality spectrum in Gene Ontology system and it can predict both generic and specific GO terms. UniGOPred was run on CAFA2 challenge target protein sequences and it is categorized within the top 10 best performing methods for the molecular function category. In addition, the performance of UniGOPred is higher compared to the baseline BLAST classifier in all categories of GO. UniGOPred predictions are compared with UniProtKB/TrEMBL database annotations as well. Furthermore, the proposed tool's ability to predict negatively associated GO terms that defines the functions that a protein does not possess, is discussed. UniGOPred annotations were also validated by case studies on PTEN protein variants experimentally and on CHD8 protein variants with literature. UniGOPred protein functional annotation system is available as an open access tool at http://cansyl.metu.edu.tr/UniGOPred.html.

Extensive complementarity between gene function prediction methods.

The number of sequenced genomes rises steadily but we still lack the knowledge about the biological roles of many genes. Automated function prediction (AFP) is thus a necessity. We hypothesized that AFP approaches that draw on distinct genome features may be useful for predicting different types of gene functions, motivating a systematic analysis of the benefits gained by obtaining and integrating such predictions. Our pipeline amalgamates 5133543 genes from 2071 genomes in a single massive analysis that evaluates five established genomic AFP methodologies. While 1227 Gene Ontology (GO) terms yielded reliable predictions, the majority of these functions were accessible to only one or two of the methods. Moreover, different methods tend to assign a GO term to non-overlapping sets of genes. Thus, inferences made by diverse genomic AFP methods display a striking complementary, both gene-wise and function-wise. Because of this, a viable integration strategy is to rely on a single most-confident prediction per gene/function, rather than enforcing agreement across multiple AFP methods. Using an information-theoretic approach, we estimate that current databases contain 29.2 bits/gene of known Escherichia coli gene functions. This can be increased by up to 5.5 bits/gene using individual AFP methods or by 11 additional bits/gene upon integration, thereby providing a highly-ranking predictor on the Critical Assessment of Function Annotation 2 community benchmark. Availability of more sequenced genomes boosts the predictive accuracy of AFP approaches and also the benefit from integrating them. The individual and integrated GO predictions for the complete set of genes are available from http://gorbi.irb.hr/ CONTACT: fran.supek@irb.hrSupplementary information: Supplementary materials are available at Bioinformatics online.

UNIPred: Unbalance-Aware Network Integration and Prediction of Protein Functions.

The proper integration of multiple sources of data and the unbalance between annotated and unannotated proteins represent two of the main issues of the automated function prediction (AFP) problem. Most of supervised and semisupervised learning algorithms for AFP proposed in literature do not jointly consider these items, with a negative impact on both sensitivity and precision performances, due to the unbalance between annotated and unannotated proteins that characterize the majority of functional classes and to the specific and complementary information content embedded in each available source of data. We propose UNIPred (unbalance-aware network integration and prediction of protein functions), an algorithm that properly combines different biomolecular networks and predicts protein functions using parametric semisupervised neural models. The algorithm explicitly takes into account the unbalance between unannotated and annotated proteins both to construct the integrated network and to predict protein annotations for each functional class. Full-genome and ontology-wide experiments with three eukaryotic model organisms show that the proposed method compares favorably with state-of-the-art learning algorithms for AFP.

The PFP and ESG protein function prediction methods in 2014: effect of database updates and ensemble approaches.

BackgroundFunctional annotation of novel proteins is one of the central problems in bioinformatics. With the ever-increasing development of genome sequencing technologies, more and more sequence information is becoming available to analyze and annotate. To achieve fast and automatic function annotation, many computational (automated) function prediction (AFP) methods have been developed. To objectively evaluate the performance of such methods on a large scale, community-wide assessment experiments have been conducted. The second round of the Critical Assessment of Function Annotation (CAFA) experiment was held in 2013–2014. Evaluation of participating groups was reported in a special interest group meeting at the Intelligent Systems in Molecular Biology (ISMB) conference in Boston in 2014. Our group participated in both CAFA1 and CAFA2 using multiple, in-house AFP methods. Here, we report benchmark results of our methods obtained in the course of preparation for CAFA2 prior to submitting function predictions for CAFA2 targets.ResultsFor CAFA2, we updated the annotation databases used by our methods, protein function prediction (PFP) and extended similarity group (ESG), and benchmarked their function prediction performances using the original (older) and updated databases. Performance evaluation for PFP with different settings and ESG are discussed. We also developed two ensemble methods that combine function predictions from six independent, sequence-based AFP methods. We further analyzed the performances of our prediction methods by enriching the predictions with prior distribution of gene ontology (GO) terms. Examples of predictions by the ensemble methods are discussed.ConclusionsUpdating the annotation database was successful, improving the Fmax prediction accuracy score for both PFP and ESG. Adding the prior distribution of GO terms did not make much improvement. Both of the ensemble methods we developed improved the average Fmax score over all individual component methods except for ESG. Our benchmark results will not only complement the overall assessment that will be done by the CAFA organizers, but also help elucidate the predictive powers of sequence-based function prediction methods in general.Electronic supplementary materialThe online version of this article (doi:10.1186/s13742-015-0083-4) contains supplementary material, which is available to authorized users.

A close look at protein function prediction evaluation protocols.

BackgroundThe recently held Critical Assessment of Function Annotation challenge (CAFA2) required its participants to submit predictions for a large number of target proteins regardless of whether they have previous annotations or not. This is in contrast to the original CAFA challenge in which participants were asked to submit predictions for proteins with no existing annotations. The CAFA2 task is more realistic, in that it more closely mimics the accumulation of annotations over time. In this study we compare these tasks in terms of their difficulty, and determine whether cross-validation provides a good estimate of performance.ResultsThe CAFA2 task is a combination of two subtasks: making predictions on annotated proteins and making predictions on previously unannotated proteins. In this study we analyze the performance of several function prediction methods in these two scenarios. Our results show that several methods (structured support vector machine, binary support vector machines and guilt-by-association methods) do not usually achieve the same level of accuracy on these two tasks as that achieved by cross-validation, and that predicting novel annotations for previously annotated proteins is a harder problem than predicting annotations for uncharacterized proteins. We also find that different methods have different performance characteristics in these tasks, and that cross-validation is not adequate at estimating performance and ranking methods.ConclusionsThese results have implications for the design of computational experiments in the area of automated function prediction and can provide useful insight for the understanding and design of future CAFA competitions.Electronic supplementary materialThe online version of this article (doi:10.1186/s13742-015-0082-5) contains supplementary material, which is available to authorized users.

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.

Think globally and solve locally: secondary memory-based network learning for automated multi-species function prediction.

BackgroundNetwork-based learning algorithms for automated function prediction (AFP) are negatively affected by the limited coverage of experimental data and limited a priori known functional annotations. As a consequence their application to model organisms is often restricted to well characterized biological processes and pathways, and their effectiveness with poorly annotated species is relatively limited. A possible solution to this problem might consist in the construction of big networks including multiple species, but this in turn poses challenging computational problems, due to the scalability limitations of existing algorithms and the main memory requirements induced by the construction of big networks. Distributed computation or the usage of big computers could in principle respond to these issues, but raises further algorithmic problems and require resources not satisfiable with simple off-the-shelf computers.ResultsWe propose a novel framework for scalable network-based learning of multi-species protein functions based on both a local implementation of existing algorithms and the adoption of innovative technologies: we solve “locally” the AFP problem, by designing “vertex-centric” implementations of network-based algorithms, but we do not give up thinking “globally” by exploiting the overall topology of the network. This is made possible by the adoption of secondary memory-based technologies that allow the efficient use of the large memory available on disks, thus overcoming the main memory limitations of modern off-the-shelf computers. This approach has been applied to the analysis of a large multi-species network including more than 300 species of bacteria and to a network with more than 200,000 proteins belonging to 13 Eukaryotic species. To our knowledge this is the first work where secondary-memory based network analysis has been applied to multi-species function prediction using biological networks with hundreds of thousands of proteins.ConclusionsThe combination of these algorithmic and technological approaches makes feasible the analysis of large multi-species networks using ordinary computers with limited speed and primary memory, and in perspective could enable the analysis of huge networks (e.g. the whole proteomes available in SwissProt), using well-equipped stand-alone machines.

The automated function prediction SIG looks back at 2013 and prepares for 2014

Abstract Contact: m.n.wass@kent.ac.uk or mark@wass.com

Protein function prediction by massive integration of evolutionary analyses and multiple data sources

BackgroundAccurate protein function annotation is a severe bottleneck when utilizing the deluge of high-throughput, next generation sequencing data. Keeping database annotations up-to-date has become a major scientific challenge that requires the development of reliable automatic predictors of protein function. The CAFA experiment provided a unique opportunity to undertake comprehensive 'blind testing' of many diverse approaches for automated function prediction. We report on the methodology we used for this challenge and on the lessons we learnt.MethodsOur method integrates into a single framework a wide variety of biological information sources, encompassing sequence, gene expression and protein-protein interaction data, as well as annotations in UniProt entries. The methodology transfers functional categories based on the results from complementary homology-based and feature-based analyses. We generated the final molecular function and biological process assignments by combining the initial predictions in a probabilistic manner, which takes into account the Gene Ontology hierarchical structure.ResultsWe propose a novel scoring function called COmbined Graph-Information Content similarity (COGIC) score for the comparison of predicted functional categories and benchmark data. We demonstrate that our integrative approach provides increased scope and accuracy over both the component methods and the naïve predictors. In line with previous studies, we find that molecular function predictions are more accurate than biological process assignments.ConclusionsOverall, the results indicate that there is considerable room for improvement in the field. It still remains for the community to invest a great deal of effort to make automated function prediction a useful and routine component in the toolbox of life scientists. As already witnessed in other areas, community-wide blind testing experiments will be pivotal in establishing standards for the evaluation of prediction accuracy, in fostering advancements and new ideas, and ultimately in recording progress.

Coupled mutation finder: A new entropy-based method quantifying phylogenetic noise for the detection of compensatory mutations

BackgroundThe detection of significant compensatory mutation signals in multiple sequence alignments (MSAs) is often complicated by noise. A challenging problem in bioinformatics is remains the separation of significant signals between two or more non-conserved residue sites from the phylogenetic noise and unrelated pair signals. Determination of these non-conserved residue sites is as important as the recognition of strictly conserved positions for understanding of the structural basis of protein functions and identification of functionally important residue regions. In this study, we developed a new method, the Coupled Mutation Finder (CMF) quantifying the phylogenetic noise for the detection of compensatory mutations.ResultsTo demonstrate the effectiveness of this method, we analyzed essential sites of two human proteins: epidermal growth factor receptor (EGFR) and glucokinase (GCK). Our results suggest that the CMF is able to separate significant compensatory mutation signals from the phylogenetic noise and unrelated pair signals. The vast majority of compensatory mutation sites found by the CMF are related to essential sites of both proteins and they are likely to affect protein stability or functionality.ConclusionsThe CMF is a new method, which includes an MSA-specific statistical model based on multiple testing procedures that quantify the error made in terms of the false discovery rate and a novel entropy-based metric to upscale BLOSUM62 dissimilar compensatory mutations. Therefore, it is a helpful tool to predict and investigate compensatory mutation sites of structural or functional importance in proteins. We suggest that the CMF could be used as a novel automated function prediction tool that is required for a better understanding of the structural basis of proteins. The CMF server is freely accessible at http://cmf.bioinf.med.uni-goettingen.de.

Analysis of protein function and its prediction from amino acid sequence

Understanding protein function is one of the keys to understanding life at the molecular level. It is also important in the context of human disease because many conditions arise as a consequence of alterations of protein function. The recent availability of relatively inexpensive sequencing technology has resulted in thousands of complete or partially sequenced genomes with millions of functionally uncharacterized proteins. Such a large volume of data, combined with the lack of high-throughput experimental assays to functionally annotate proteins, attributes to the growing importance of automated function prediction. Here, we study proteins annotated by Gene Ontology (GO) terms and estimate the accuracy of functional transfer from protein sequence only. We find that the transfer of GO terms by pairwise sequence alignments is only moderately accurate, showing a surprisingly small influence of sequence identity (SID) in a broad range (30-100%). We developed and evaluated a new predictor of protein function, functional annotator (FANN), from amino acid sequence. The predictor exploits a multioutput neural network framework which is well suited to simultaneously modeling dependencies between functional terms. Experiments provide evidence that FANN-GO (predictor of GO terms; available from http://www.informatics.indiana.edu/predrag) outperforms standard methods such as transfer by global or local SID as well as GOtcha, a method that incorporates the structure of GO.

Application of maximin correlation analysis to classifying protein environments for function prediction

More and more protein structures are being discovered, but most of these still have little functional information. Based on the assumption that structural resemblance would lead to functional similarity, researchers computationally compare a new structure with functionally annotated structures, for high-throughput function prediction. The effectiveness of this approach depends critically upon the quality of comparison. In particular, robust classification often becomes difficult when a function class is an aggregate of multiple subclasses, as is the case with protein annotations. For such multiple-subclass classification problems, an optimal method termed the maximin correlation analysis (MCA) was proposed. However, MCA has never been applied to automated protein function prediction although MCA can minimize the misclassification risk in the correlation-based nearest neighbor classification, thus increasing classification accuracy. In this article, we apply MCA to classifying three-dimensional protein local environment data derived from a subset of the protein data bank (PDB). In our framework, the MCA-based classifier outperformed the compared alternatives by 7–19% and 6–27% in terms of average sensitivity and specificity, respectively. Given that correlation-based similarity measures have been widely used for mining protein data, we expect that MCA would be employed to enhance other types of automated function prediction methods.

An overview of in silico protein function prediction

As the protein databases continue to expand at an exponential rate, fed by daily uploads from multiple large scale genomic and metagenomic projects, the problem of assigning a function to each new protein has become the focus of significant research interest in recent times. Herein, we review the most recent advances in the field of automated function prediction (AFP). We begin by defining what is meant by biological "function" and the means of describing such functions using standardised machine readable ontologies. We then focus on the various function-prediction programs available, both sequence and structure based, and outline their associated strengths and weaknesses. Finally, we conclude with a brief overview of the future challenges and outstanding questions in the field, which still remain unanswered.

An efficient strategy for extensive integration of diverse biological data for protein function prediction

With the increasing availability of diverse biological information, protein function prediction approaches have converged towards integration of heterogeneous data. Many adapted existing techniques, such as machine-learning and probabilistic methods, which have proven successful on specific data types. However, the impact of these approaches is hindered by a couple of factors. First, there is little comparison between existing approaches. This is in part due to a divergence in the focus adopted by different works, which makes comparison difficult or even fuzzy. Second, there seems to be over-emphasis on the use of computationally demanding machine-learning methods, which runs counter to the surge in biological data. Analogous to the success of BLAST for sequence homology search, we believe that the ability to tap escalating quantity, quality and diversity of biological data is crucial to the success of automated function prediction as a useful instrument for the advancement of proteomic research. We address these problems by: (1) providing useful comparison between some prominent methods; (2) proposing Integrated Weighted Averaging (IWA)--a scalable, efficient and flexible function prediction framework that integrates diverse information using simple weighting strategies and a local prediction method. The simplicity of the approach makes it possible to make predictions based on on-the-fly information fusion. In addition to its greater efficiency, IWA performs exceptionally well against existing approaches. In the presence of cross-genome information, which is overwhelming for existing approaches, IWA makes even better predictions. We also demonstrate the significance of appropriate weighting strategies in data integration.