An SVM-based system for predicting protein subnuclear localizations.
ABSTRACT: BACKGROUND: The large gap between the number of protein sequences in databases and the number of functionally characterized proteins calls for the development of a fast computational tool for the prediction of subnuclear and subcellular localizations generally applicable to protein sequences. The information on localization may reveal the molecular function of novel proteins, in addition to providing insight on the biological pathways in which they function. The bulk of past work has been focused on protein subcellular localizations. Furthermore, no specific tool has been dedicated to prediction at the subnuclear level, despite its high importance. In order to design a suitable predictive system, the extraction of subtle sequence signals that can discriminate among proteins with different subnuclear localizations is the key. RESULTS: New kernel functions used in a support vector machine (SVM) learning model are introduced for the measurement of sequence similarity. The k-peptide vectors are first mapped by a matrix of high-scored pairs of k-peptides which are measured by BLOSUM62 scores. The kernels, measuring the similarity for sequences, are then defined on the mapped vectors. By combining these new encoding methods, a multi-class classification system for the prediction of protein subnuclear localizations is established for the first time. The performance of the system is evaluated with a set of proteins collected in the Nuclear Protein Database (NPD). The overall accuracy of prediction for 6 localizations is about 50% (vs. random prediction 16.7%) for single localization proteins in the leave-one-out cross-validation; and 65% for an independent set of multi-localization proteins. This integrated system can be accessed at http://array.bioengr.uic.edu/subnuclear.htm. CONCLUSION: The integrated system benefits from the combination of predictions from several SVMs based on selected encoding methods. Finally, the predictive power of the system is expected to improve as more proteins with known subnuclear localizations become available.
Project description:BACKGROUND: The accomplishment of the various genome sequencing projects resulted in accumulation of massive amount of gene sequence information. This calls for a large-scale computational method for predicting protein localization from sequence. The protein localization can provide valuable information about its molecular function, as well as the biological pathway in which it participates. The prediction of localization of a protein at subnuclear level is a challenging task. In our previous work we proposed an SVM-based system using protein sequence information for this prediction task. In this work, we assess protein similarity with Gene Ontology (GO) and then improve the performance of the system by adding a module of nearest neighbor classifier using a similarity measure derived from the GO annotation terms for protein sequences. RESULTS: The performance of the new system proposed here was compared with our previous system using a set of proteins resided within 6 localizations collected from the Nuclear Protein Database (NPD). The overall MCC (accuracy) is elevated from 0.284 (50.0%) to 0.519 (66.5%) for single-localization proteins in leave-one-out cross-validation; and from 0.420 (65.2%) to 0.541 (65.2%) for an independent set of multi-localization proteins. The new system is available at http://array.bioengr.uic.edu/subnuclear.htm. CONCLUSION: The prediction of protein subnuclear localizations can be largely influenced by various definitions of similarity for a pair of proteins based on different similarity measures of GO terms. Using the sum of similarity scores over the matched GO term pairs for two proteins as the similarity definition produced the best predictive outcome. Substantial improvement in predicting protein subnuclear localizations has been achieved by combining Gene Ontology with sequence information.
Project description:Computational methods have been widely used for the prediction of protein subcellular localization. However, these predictions are rarely validated experimentally and as a result remain questionable. Therefore, experimental validation of the predicted localizations is needed to assess the accuracy of predictions so that such methods can be confidently used to annotate the proteins of unknown localization. Previously, we published a method called ngLOC that predicts the localization of proteins targeted to ten different subcellular organelles. In this short report, we describe the accuracy of these predictions using experimental validations.We have experimentally validated the predicted subcellular localizations of 114 human proteins corresponding to nine different organelles in normal breast and breast cancer cell lines using live cell imaging/confocal microscopy. Target genes were cloned into expression vectors as GFP fusions and cotransfected with RFP-tagged organelle-specific gene marker into normal breast epithelial and breast cancer cell lines. Subcellular localization of each target protein is confirmed by colocalization with a co-expressed organelle-specific protein marker. Our results showed that about 82.5% of the predicted subcellular localizations coincided with the experimentally validated localizations. The highest agreement was found in the endoplasmic reticulum proteins, while the cytoplasmic location showed the least concordance. With the exclusion of cytoplasmic location, the average prediction accuracy increased to 90.4%. In addition, there was no difference observed in the protein subcellular localization between normal and cancer breast cell lines.The experimentally validated accuracy of ngLOC method with (82.5%) or without cytoplasmic location (90.4%) nears the prediction accuracy of 89%. These results demonstrate that the ngLOC method can be very useful for large-scale annotation of the unknown subcellular localization of proteins.
Project description:BACKGROUND: Knowledge of subcellular localization of proteins is crucial to proteomics, drug target discovery and systems biology since localization and biological function are highly correlated. In recent years, numerous computational prediction methods have been developed. Nevertheless, there is still a need for prediction methods that show more robustness and higher accuracy. RESULTS: We extended our previous MultiLoc predictor by incorporating phylogenetic profiles and Gene Ontology terms. Two different datasets were used for training the system, resulting in two versions of this high-accuracy prediction method. One version is specialized for globular proteins and predicts up to five localizations, whereas a second version covers all eleven main eukaryotic subcellular localizations. In a benchmark study with five localizations, MultiLoc2 performs considerably better than other methods for animal and plant proteins and comparably for fungal proteins. Furthermore, MultiLoc2 performs clearly better when using a second dataset that extends the benchmark study to all eleven main eukaryotic subcellular localizations. CONCLUSION: MultiLoc2 is an extensive high-performance subcellular protein localization prediction system. By incorporating phylogenetic profiles and Gene Ontology terms MultiLoc2 yields higher accuracies compared to its previous version. Moreover, it outperforms other prediction systems in two benchmarks studies. MultiLoc2 is available as user-friendly and free web-service, available at: http://www-bs.informatik.uni-tuebingen.de/Services/MultiLoc2.
Project description:With the rapid increase of protein sequences in the post-genomic age, it is challenging to develop accurate and automated methods for reliably and quickly predicting their subcellular localizations. Till now, many efforts have been tried, but most of which used only a single algorithm. In this paper, we proposed an ensemble classifier of KNN (k-nearest neighbor) and SVM (support vector machine) algorithms to predict the subcellular localization of eukaryotic proteins based on a voting system. The overall prediction accuracies by the one-versus-one strategy are 78.17%, 89.94% and 75.55% for three benchmark datasets of eukaryotic proteins. The improved prediction accuracies reveal that GO annotations and hydrophobicity of amino acids help to predict subcellular locations of eukaryotic proteins.
Project description:BACKGROUND: Gene Ontology (GO) annotation, which describes the function of genes and gene products across species, has recently been used to predict protein subcellular and subnuclear localization. Existing GO-based prediction methods for protein subcellular localization use the known accession numbers of query proteins to obtain their annotated GO terms. An accurate prediction method for predicting subcellular localization of novel proteins without known accession numbers, using only the input sequence, is worth developing. RESULTS: This study proposes an efficient sequence-based method (named ProLoc-GO) by mining informative GO terms for predicting protein subcellular localization. For each protein, BLAST is used to obtain a homology with a known accession number to the protein for retrieving the GO annotation. A large number n of all annotated GO terms that have ever appeared are then obtained from a large set of training proteins. A novel genetic algorithm based method (named GOmining) combined with a classifier of support vector machine (SVM) is proposed to simultaneously identify a small number m out of the n GO terms as input features to SVM, where m <<n. The m informative GO terms contain the essential GO terms annotating subcellular compartments such as GO:0005634 (Nucleus), GO:0005737 (Cytoplasm) and GO:0005856 (Cytoskeleton). Two existing data sets SCL12 (human protein with 12 locations) and SCL16 (Eukaryotic proteins with 16 locations) with <25% sequence identity are used to evaluate ProLoc-GO which has been implemented by using a single SVM classifier with the m = 44 and m = 60 informative GO terms, respectively. ProLoc-GO using input sequences yields test accuracies of 88.1% and 83.3% for SCL12 and SCL16, respectively, which are significantly better than the SVM-based methods, which achieve < 35% test accuracies using amino acid composition (AAC) with acid pairs and AAC with dipedtide composition. For comparison, ProLoc-GO using known accession numbers of query proteins yields test accuracies of 90.6% and 85.7%, which is also better than Hum-PLoc (85.0%) and Euk-OET-PLoc (83.7%) using ensemble classifiers with hybridization of GO terms and amphiphilic pseudo amino acid composition for SCL12 and SCL16, respectively. CONCLUSION: The growth of Gene Ontology in size and popularity has increased the effectiveness of GO-based features. GOmining can serve as a tool for selecting informative GO terms in solving sequence-based prediction problems. The prediction system using ProLoc-GO with input sequences of query proteins for protein subcellular localization has been implemented (see Availability).
Project description:BACKGROUND: Most proteins have evolved in specific cellular compartments that limit their functions and potential interactions. On the other hand, motifs define amino acid arrangements conserved between protein family members and represent powerful tools for assigning function to protein sequences. The ideal motif would identify all members of a protein family but in practice many motifs identify both family members and unrelated proteins, referred to as True Positive (TP) and False Positive (FP) sequences, respectively. RESULTS: To address the relationship between protein motifs, protein function and cellular localization, we systematically assigned subcellular localization data to motif sequences from the comprehensive PROSITE sequence motif database. Using this data we analyzed relationships between localization and function. We find that TPs and FPs have a strong tendency to localize in different compartments. When multiple localizations are considered, TPs are usually distributed between related cellular compartments. We also identified cases where FPs are concentrated in particular subcellular regions, indicating possible functional or evolutionary relationships with TP sequences of the same motif. CONCLUSIONS: Our findings suggest that the systematic examination of subcellular localization has the potential to uncover evolutionary and functional relationships between motif-containing sequences. We believe that this type of analysis complements existing motif annotations and could aid in their interpretation. Our results shed light on the evolution of cellular organelles and potentially establish the basis for new subcellular localization and function prediction algorithms.
Project description:Proteins targeting the same subcellular localization tend to participate in mutual protein-protein interactions (PPIs) and are often functionally associated. Here, we investigated the relationship between disease-associated proteins and their subcellular localizations, based on the assumption that protein pairs associated with phenotypically similar diseases are more likely to be connected via subcellular localization. The spatial constraints from subcellular localization significantly strengthened the disease associations of the proteins connected by subcellular localizations. In particular, certain disease types were more prevalent in specific subcellular localizations. We analyzed the enrichment of disease phenotypes within subcellular localizations, and found that there exists a significant correlation between disease classes and subcellular localizations. Furthermore, we found that two diseases displayed high comorbidity when disease-associated proteins were connected via subcellular localization. We newly explained 7584 disease pairs by using the context of protein subcellular localization, which had not been identified using shared genes or PPIs only. Our result establishes a direct correlation between protein subcellular localization and disease association, and helps to understand the mechanism of human disease progression.
Project description:BACKGROUND: Proteins that interact in vivo tend to reside within the same or "adjacent" subcellular compartments. This observation provides opportunities to reveal protein subcellular localization in the context of the protein-protein interaction (PPI) network. However, so far, only a few efforts based on heuristic rules have been made in this regard. RESULTS: We systematically and quantitatively validate the hypothesis that proteins physically interacting with each other probably share at least one common subcellular localization. With the result, for the first time, four graph-based semi-supervised learning algorithms, Majority, ?2-score, GenMultiCut and FunFlow originally proposed for protein function prediction, are introduced to assign "multiplex localization" to proteins. We analyze these approaches by performing a large-scale cross validation on a Saccharomyces cerevisiae proteome compiled from BioGRID and comparing their predictions for 22 protein subcellular localizations. Furthermore, we build an ensemble classifier to associate 529 unlabeled and 137 ambiguously-annotated proteins with subcellular localizations, most of which have been verified in the previous experimental studies. CONCLUSIONS: Physical interaction of proteins has actually provided an essential clue for their co-localization. Compared to the local approaches, the global algorithms consistently achieve a superior performance.
Project description:Despite the growing volume of experimentally validated knowledge about the subcellular localization of plant proteins, a well performing in silico prediction tool is still a necessity. Existing tools, which employ information derived from protein sequence alone, offer limited accuracy and/or rely on full sequence availability. We explored whether gene expression profiling data can be harnessed to enhance prediction performance. To achieve this, we trained several support vector machines to predict the subcellular localization of Arabidopsis thaliana proteins using sequence derived information, expression behavior, or a combination of these data and compared their predictive performance through a cross-validation test. We show that gene expression carries information about the subcellular localization not available in sequence information, yielding dramatic benefits for plastid localization prediction, and some notable improvements for other compartments such as the mitochondrion, the Golgi, and the plasma membrane. Based on these results, we constructed a novel subcellular localization prediction engine, SLocX, combining gene expression profiling data with protein sequence-based information. We then validated the results of this engine using an independent test set of annotated proteins and a transient expression of GFP fusion proteins. Here, we present the prediction framework and a website of predicted localizations for Arabidopsis. The relatively good accuracy of our prediction engine, even in cases where only partial protein sequence is available (e.g., in sequences lacking the N-terminal region), offers a promising opportunity for similar application to non-sequenced or poorly annotated plant species. Although the prediction scope of our method is currently limited by the availability of expression information on the ATH1 array, we believe that the advances in measuring gene expression technology will make our method applicable for all Arabidopsis proteins.
Project description:BACKGROUND: Identification of a bacterial protein's subcellular localization (SCL) is important for genome annotation, function prediction and drug or vaccine target identification. Subcellular fractionation techniques combined with recent proteomics technology permits the identification of large numbers of proteins from distinct bacterial compartments. However, the fractionation of a complex structure like the cell into several subcellular compartments is not a trivial task. Contamination from other compartments may occur, and some proteins may reside in multiple localizations. New computational methods have been reported over the past few years that now permit much more accurate, genome-wide analysis of the SCL of protein sequences deduced from genomes. There is a need to compare such computational methods with laboratory proteomics approaches to identify the most effective current approach for genome-wide localization characterization and annotation. RESULTS: In this study, ten subcellular proteome analyses of bacterial compartments were reviewed. PSORTb version 2.0 was used to computationally predict the localization of proteins reported in these publications, and these computational predictions were then compared to the localizations determined by the proteomics study. By using a combined approach, we were able to identify a number of contaminants and proteins with dual localizations, and were able to more accurately identify membrane subproteomes. Our results allowed us to estimate the precision level of laboratory subproteome studies and we show here that, on average, recent high-precision computational methods such as PSORTb now have a lower error rate than laboratory methods. CONCLUSION: We have performed the first focused comparison of genome-wide proteomic and computational methods for subcellular localization identification, and show that computational methods have now attained a level of precision that is exceeding that of high-throughput laboratory approaches. We note that analysis of all cellular fractions collectively is required to effectively provide localization information from laboratory studies, and we propose an overall approach to genome-wide subcellular localization characterization that capitalizes on the complementary nature of current laboratory and computational methods.