A large-scale evaluation of computational protein function prediction.
Ontology highlight
ABSTRACT:
SUBMITTER: Radivojac P
PROVIDER: S-EPMC3584181 | biostudies-literature | 2013 Mar
REPOSITORIES: biostudies-literature
Publications
A large-scale evaluation of computational protein function prediction.
Radivojac Predrag P Clark Wyatt T WT Oron Tal Ronnen TR Schnoes Alexandra M AM Wittkop Tobias T Sokolov Artem A Graim Kiley K Funk Christopher C Verspoor Karin K Ben-Hur Asa A Pandey Gaurav G Yunes Jeffrey M JM Talwalkar Ameet S AS Repo Susanna S Souza Michael L ML Piovesan Damiano D Casadio Rita R Wang Zheng Z Cheng Jianlin J Fang Hai H Gough Julian J Koskinen Patrik P Törönen Petri P Nokso-Koivisto Jussi J Holm Liisa L Cozzetto Domenico D Buchan Daniel W A DW Bryson Kevin K Jones David T DT Limaye Bhakti B Inamdar Harshal H Datta Avik A Manjari Sunitha K SK Joshi Rajendra R Chitale Meghana M Kihara Daisuke D Lisewski Andreas M AM Erdin Serkan S Venner Eric E Lichtarge Olivier O Rentzsch Robert R Yang Haixuan H Romero Alfonso E AE Bhat Prajwal P Paccanaro Alberto A Hamp Tobias T Kaßner Rebecca R Seemayer Stefan S Vicedo Esmeralda E Schaefer Christian C Achten Dominik D Auer Florian F Boehm Ariane A Braun Tatjana T Hecht Maximilian M Heron Mark M Hönigschmid Peter P Hopf Thomas A TA Kaufmann Stefanie S Kiening Michael M Krompass Denis D Landerer Cedric C Mahlich Yannick Y Roos Manfred M Björne Jari J Salakoski Tapio T Wong Andrew A Shatkay Hagit H Gatzmann Fanny F Sommer Ingolf I Wass Mark N MN Sternberg Michael J E MJ Škunca Nives N Supek Fran F Bošnjak Matko M Panov Panče P Džeroski Sašo S Šmuc Tomislav T Kourmpetis Yiannis A I YA van Dijk Aalt D J AD ter Braak Cajo J F CJ Zhou Yuanpeng Y Gong Qingtian Q Dong Xinran X Tian Weidong W Falda Marco M Fontana Paolo P Lavezzo Enrico E Di Camillo Barbara B Toppo Stefano S Lan Liang L Djuric Nemanja N Guo Yuhong Y Vucetic Slobodan S Bairoch Amos A Linial Michal M Babbitt Patricia C PC Brenner Steven E SE Orengo Christine C Rost Burkhard B Mooney Sean D SD Friedberg Iddo I
Nature methods 20130127 3
Automated annotation of protein function is challenging. As the number of sequenced genomes rapidly grows, the overwhelming majority of protein products can only be annotated computationally. If computational predictions are to be relied upon, it is crucial that the accuracy of these methods be high. Here we report the results from the first large-scale community-based critical assessment of protein function annotation (CAFA) experiment. Fifty-four methods representing the state of the art for p ...[more]
Similar Datasets
Project description:Heterogeneous ensembles are an effective approach in scenarios where the ideal data type and/or individual predictor are unclear for a given problem. These ensembles have shown promise for protein function prediction (PFP), but their ability to improve PFP at a large scale is unclear. The overall goal of this study is to critically assess this ability of a variety of heterogeneous ensemble methods across a multitude of functional terms, proteins and organisms. Our results show that these methods, especially Stacking using Logistic Regression, indeed produce more accurate predictions for a variety of Gene Ontology terms differing in size and specificity. To enable the application of these methods to other related problems, we have publicly shared the HPC-enabled code underlying this work as LargeGOPred ( https://github.com/GauravPandeyLab/LargeGOPred).
| S-EPMC6221071 | biostudies-literature
Project description:MotivationAutomated function prediction (AFP) of proteins is a large-scale multi-label classification problem. Two limitations of most network-based methods for AFP are (i) a single model must be trained for each species and (ii) protein sequence information is totally ignored. These limitations cause weaker performance than sequence-based methods. Thus, the challenge is how to develop a powerful network-based method for AFP to overcome these limitations.ResultsWe propose DeepGraphGO, an end-to-end, multispecies graph neural network-based method for AFP, which makes the most of both protein sequence and high-order protein network information. Our multispecies strategy allows one single model to be trained for all species, indicating a larger number of training samples than existing methods. Extensive experiments with a large-scale dataset show that DeepGraphGO outperforms a number of competing state-of-the-art methods significantly, including DeepGOPlus and three representative network-based methods: GeneMANIA, deepNF and clusDCA. We further confirm the effectiveness of our multispecies strategy and the advantage of DeepGraphGO over so-called difficult proteins. Finally, we integrate DeepGraphGO into the state-of-the-art ensemble method, NetGO, as a component and achieve a further performance improvement.Availability and implementationhttps://github.com/yourh/DeepGraphGO.Supplementary informationSupplementary data are available at Bioinformatics online.
| S-EPMC8294856 | biostudies-literature
Project description:Automated function prediction (AFP) of proteins is of great significance in biology. AFP can be regarded as a problem of the large-scale multi-label classification where a protein can be associated with multiple gene ontology terms as its labels. Based on our GOLabeler-a state-of-the-art method for the third critical assessment of functional annotation (CAFA3), in this paper we propose NetGO, a web server that is able to further improve the performance of the large-scale AFP by incorporating massive protein-protein network information. Specifically, the advantages of NetGO are threefold in using network information: (i) NetGO relies on a powerful learning to rank framework from machine learning to effectively integrate both sequence and network information of proteins; (ii) NetGO uses the massive network information of all species (>2000) in STRING (other than only some specific species) and (iii) NetGO still can use network information to annotate a protein by homology transfer, even if it is not contained in STRING. Separating training and testing data with the same time-delayed settings of CAFA, we comprehensively examined the performance of NetGO. Experimental results have clearly demonstrated that NetGO significantly outperforms GOLabeler and other competing methods. The NetGO web server is freely available at http://issubmission.sjtu.edu.cn/netgo/.
| S-EPMC6602452 | biostudies-literature
Project description:BackgroundThe prediction of protein-protein interactions is an important step toward the elucidation of protein functions and the understanding of the molecular mechanisms inside the cell. While experimental methods for identifying these interactions remain costly and often noisy, the increasing quantity of solved 3D protein structures suggests that in silico methods to predict interactions between two protein structures will play an increasingly important role in screening candidate interacting pairs. Approaches using the knowledge of the structure are presumably more accurate than those based on sequence only. Approaches based on docking protein structures solve a variant of this problem, but these methods remain very computationally intensive and will not scale in the near future to the detection of interactions at the level of an interactome, involving millions of candidate pairs of proteins.ResultsHere, we describe a computational method to predict efficiently in silico whether two protein structures interact. This yes/no question is presumably easier to answer than the standard protein docking question, "How do these two protein structures interact?" Our approach is to discriminate between interacting and non-interacting protein pairs using a statistical pattern recognition method known as a support vector machine (SVM). We demonstrate that our structure-based method performs well on this task and scales well to the size of an interactome.ConclusionsThe use of structure information for the prediction of protein interaction yields significantly better performance than other sequence-based methods. Among structure-based classifiers, the SVM algorithm, combined with the metric learning pairwise kernel and the MAMMOTH kernel, performs best in our experiments.
| S-EPMC2845582 | biostudies-literature
Project description:Although the identification of protein interactions by high-throughput methods progresses at a fast pace, "interactome" datasets still suffer from high rates of false positives and low coverage. To map the interactome of any organism, this unit presents a computational framework to predict protein-protein or gene-gene interactions utilizing experimentally determined evidence of structural complexes, atomic details of binding interfaces and evolutionary conservation.
| S-EPMC3920302 | biostudies-literature
Project description:Sequence database searches followed by homology-based function transfer form one of the oldest and most popular approaches for predicting protein functions, such as Gene Ontology (GO) terms. These searches are also a critical component in most state-of-the-art machine learning and deep learning-based protein function predictors. Although sequence search tools are the basis of homology-based protein function prediction, previous studies have scarcely explored how to select the optimal sequence search tools and configure their parameters to achieve the best function prediction. In this paper, we evaluate the effect of using different options from among popular search tools, as well as the impacts of search parameters, on protein function prediction. When predicting GO terms on a large benchmark dataset, we found that BLASTp and MMseqs2 consistently exceed the performance of other tools, including DIAMOND-one of the most popular tools for function prediction-under default search parameters. However, with the correct parameter settings, DIAMOND can perform comparably to BLASTp and MMseqs2 in function prediction. Additionally, we developed a new scoring function to derive GO prediction from homologous hits that consistently outperform previously proposed scoring functions. These findings enable the improvement of almost all protein function prediction algorithms with a few easily implementable changes in their sequence homolog-based component. This study emphasizes the critical role of search parameter settings in homology-based function transfer and should have an important contribution to the development of future protein function prediction algorithms.
| S-EPMC11262835 | biostudies-literature
Project description:Sequence database searches followed by homology-based function transfer form one of the oldest and most popular approaches for predicting protein functions, such as Gene Ontology (GO) terms. Although sequence search tools are the basis of homology-based protein function prediction, previous studies have scarcely explored how to select the optimal sequence search tools and configure their parameters to achieve the best function prediction. In this paper, we evaluate the effect of using different options from among popular search tools, as well as the impacts of search parameters, on protein function prediction. When predicting GO terms on a large benchmark dataset, we found that BLASTp and MMseqs2 consistently exceed the performance of other tools, including DIAMOND - one of the most popular tools for function prediction - under default search parameters. However, with the correct parameter settings, DIAMOND can perform comparably to BLASTp and MMseqs2 in function prediction. This study emphasizes the critical role of search parameter settings in homology-based function transfer.
| S-EPMC10680702 | biostudies-literature
Project description:ImportanceWhether protein risk scores derived from a single plasma sample could be useful for risk assessment for atherosclerotic cardiovascular disease (ASCVD), in conjunction with clinical risk factors and polygenic risk scores, is uncertain.ObjectiveTo develop protein risk scores for ASCVD risk prediction and compare them to clinical risk factors and polygenic risk scores in primary and secondary event populations.Design, setting, and participantsThe primary analysis was a retrospective study of primary events among 13 540 individuals in Iceland (aged 40-75 years) with proteomics data and no history of major ASCVD events at recruitment (study duration, August 23, 2000 until October 26, 2006; follow-up through 2018). We also analyzed a secondary event population from a randomized, double-blind lipid-lowering clinical trial (2013-2016), consisting of individuals with stable ASCVD receiving statin therapy and for whom proteomic data were available for 6791 individuals.ExposuresProtein risk scores (based on 4963 plasma protein levels and developed in a training set in the primary event population); polygenic risk scores for coronary artery disease and stroke; and clinical risk factors that included age, sex, statin use, hypertension treatment, type 2 diabetes, body mass index, and smoking status at the time of plasma sampling.Main outcomes and measuresOutcomes were composites of myocardial infarction, stroke, and coronary heart disease death or cardiovascular death. Performance was evaluated using Cox survival models and measures of discrimination and reclassification that accounted for the competing risk of non-ASCVD death.ResultsIn the primary event population test set (4018 individuals [59.0% women]; 465 events; median follow-up, 15.8 years), the protein risk score had a hazard ratio (HR) of 1.93 per SD (95% CI, 1.75 to 2.13). Addition of protein risk score and polygenic risk scores significantly increased the C index when added to a clinical risk factor model (C index change, 0.022 [95% CI, 0.007 to 0.038]). Addition of the protein risk score alone to a clinical risk factor model also led to a significantly increased C index (difference, 0.014 [95% CI, 0.002 to 0.028]). Among White individuals in the secondary event population (6307 participants; 432 events; median follow-up, 2.2 years), the protein risk score had an HR of 1.62 per SD (95% CI, 1.48 to 1.79) and significantly increased C index when added to a clinical risk factor model (C index change, 0.026 [95% CI, 0.011 to 0.042]). The protein risk score was significantly associated with major adverse cardiovascular events among individuals of African and Asian ancestries in the secondary event population.Conclusions and relevanceA protein risk score was significantly associated with ASCVD events in primary and secondary event populations. When added to clinical risk factors, the protein risk score and polygenic risk score both provided statistically significant but modest improvement in discrimination.
| S-EPMC10445198 | biostudies-literature
Project description:BackgroundDespite increased investment in pharmaceutical research and development, fewer and fewer new drugs are entering the marketplace. This has prompted studies in repurposing existing drugs for use against diseases with unmet medical needs. A popular approach is to develop a classification model based on drugs with and without a desired therapeutic effect. For this approach to be statistically sound, it requires a large number of drugs in both classes. However, given few or no approved drugs for the diseases of highest medical urgency and interest, different strategies need to be investigated.ResultsWe developed a computational method termed "drug-protein interaction-based repurposing" (DPIR) that is potentially applicable to diseases with very few approved drugs. The method, based on genome-wide drug-protein interaction information and Bayesian statistics, first identifies drug-protein interactions associated with a desired therapeutic effect. Then, it uses key drug-protein interactions to score other drugs for their potential to have the same therapeutic effect.ConclusionsDetailed cross-validation studies using United States Food and Drug Administration-approved drugs for hypertension, human immunodeficiency virus, and malaria indicated that DPIR provides robust predictions. It achieves high levels of enrichment of drugs approved for a disease even with models developed based on a single drug known to treat the disease. Analysis of our model predictions also indicated that the method is potentially useful for understanding molecular mechanisms of drug action and for identifying protein targets that may potentiate the desired therapeutic effects of other drugs (combination therapies).
| S-EPMC4079911 | biostudies-literature
Project description:With the explosive growth of protein sequences, large-scale automated protein function prediction (AFP) is becoming challenging. A protein is usually associated with dozens of gene ontology (GO) terms. Therefore, AFP is regarded as a problem of large-scale multi-label classification. Under the learning to rank (LTR) framework, our previous NetGO tool integrated massive networks and multi-type information about protein sequences to achieve good performance by dealing with all possible GO terms (>44 000). In this work, we propose the updated version as NetGO 2.0, which further improves the performance of large-scale AFP. NetGO 2.0 also incorporates literature information by logistic regression and deep sequence information by recurrent neural network (RNN) into the framework. We generate datasets following the critical assessment of functional annotation (CAFA) protocol. Experiment results show that NetGO 2.0 outperformed NetGO significantly in biological process ontology (BPO) and cellular component ontology (CCO). In particular, NetGO 2.0 achieved a 12.6% improvement over NetGO in terms of area under precision-recall curve (AUPR) in BPO and around 2.6% in terms of $\mathbf {F_{max}}$ in CCO. These results demonstrate the benefits of incorporating text and deep sequence information for the functional annotation of BPO and CCO. The NetGO 2.0 web server is freely available at http://issubmission.sjtu.edu.cn/ng2/.
| S-EPMC8262706 | biostudies-literature