Predicting subcellular localization of proteins using machine-learned classifiers

MOTIVATION: Identifying the destination or localization of proteins is key to understanding their function and facilitating their purification. A number of existing computational prediction methods are based on sequence analysis. However, these methods are limited in scope, accuracy and most particularly breadth of coverage. Rather than using sequence information alone, we have explored the use of database text annotations from homologs and machine learning to substantially improve the prediction of subcellular location. RESULTS: We have constructed five machine-learning classifiers for predicting subcellular localization of proteins from animals, plants, fungi, Gram-negative bacteria and Gram-positive bacteria, which are 81% accurate for fungi and 92-94% accurate for the other four categories. These are the most accurate subcellular predictors across the widest set of organisms ever published. Our predictors are part of the Proteome Analyst web-service.     

Predicting protein subcellular localization by pseudo amino acid composition with a segment-weighted and features-combined approach

Information of protein subcellular location plays an important role in molecular cell biology. Prediction of the subcellular location of proteins will help to understand their functions and interactions. In this paper, a different mode of pseudo amino acid composition was proposed to represent protein samples for predicting their subcellular localization via the following procedures: based on the optimal splice site of each protein sequence, we divided a sequence into sorting signal part and mature protein part, and extracted sequence features from each part separately. Then, the combined features were fed into the SVM classifier to perform the prediction. By the jackknife test on a benchmark dataset in which none of proteins included has more than 90% pairwise sequence identity to any other, the overall accuracies achieved by the method are 94.5% and 90.3% for prokaryotic and eukaryotic proteins, respectively. The results indicate that the prediction quality by our method is quite satisfactory. It is anticipated that the current method may serve as an alternative approach to the existing prediction methods.     

The Molecular Evolution of the Small Heat-Shock Proteins in Plants

The small heat-shock proteins have undergone a tremendous diversification in plants; whereas only a single small heat-shock protein is found in fungi and many animals, over 20 different small heat-shock proteins are found in higher plants. The small heat-shock proteins in plants have diversified in both sequence and cellular localization and are encoded by at least five gene families. In this study, 44 small heat-shock protein DNA and amino acid sequences were examined, using both phylogenetic analysis and analysis of nucleotide substitution patterns to elucidate the evolutionary history of the small heat-shock proteins. The phylogenetic relationships of the small heat-shock proteins, estimated using parsimony and distance methods, reveal that gene duplication, sequence divergence and gene conversion have all played a role in the evolution of the small heat-shock proteins. Analysis of nonsynonymous substitutions and conservative and radical replacement substitutions (in relation to hydrophobicity) indicates that the small heat-shock protein gene families are evolving at different rates. This suggests that the small heat-shock proteins may have diversified in function as well as in sequence and cellular localization.     

SCL-Epred: a generalised de novo eukaryotic protein subcellular localisation predictor

Knowledge of the subcellular location of a protein provides valuable information about its function, possible interaction with other proteins and drug targetability, among other things. The experimental determination of a protein’s location in the cell is expensive, time consuming and open to human error. Fast and accurate predictors of subcellular location have an important role to play if the abundance of sequence data which is now available is to be fully exploited. In the post-genomic era, genomes in many diverse organisms are available. Many of these organisms are important in human and veterinary disease and fall outside of the well-studied plant, animal and fungi groups. We have developed a general eukaryotic subcellular localisation predictor (SCL-Epred) which predicts the location of eukaryotic proteins into three classes which are important, in particular, for determining the drug targetability of a protein—secreted proteins, membrane proteins and proteins that are neither secreted nor membrane. The algorithm powering SCL-Epred is a N-to-1 neural network and is trained on very large non-redundant sets of protein sequences. SCL-Epred performs well on training data achieving a Q of 86 % and a generalised correlation of 0.75 when tested in tenfold cross-validation on a set of 15,202 redundancy reduced protein sequences. The three class accuracy of SCL-Epred and LocTree2, and in particular a consensus predictor comprising both methods, surpasses that of other widely used predictors when benchmarked using a large redundancy reduced independent test set of 562 proteins. SCL-Epred is publicly available at http://distillf.ucd.ie/distill/.     

ngLOC: software and web server for predicting protein subcellular localization in prokaryotes and eukaryotes

ABSTRACT: BACKGROUND: Understanding protein subcellular localization is a necessary component toward understanding the overall function of a protein. Numerous computational methods have been published over the past decade, with varying degrees of success. Despite the large number of published methods in this area, only a small fraction of them are available for researchers to use in their own studies. Of those that are available, many are limited by predicting only a small number of major organelles in the cell. Additionally, the majority of methods predict only a single location, even though it is known that a large fraction of the proteins in eukaryotic species shuttle between locations to carry out their function. FINDINGS: We present a software package and a web server for predicting subcellular localization of protein sequences based on the ngLOC method. ngLOC is an n-gram-based Bayesian classifier that predicts subcellular localization of proteins both in prokaryotes and eukaryotes. The overall prediction accuracy varies from 89.8% to 91.4% across species. This program can predict 11 distinct locations each in plant and animal species. ngLOC also predicts 4 and 5 distinct locations on gram-positive and gram-negative bacterial datasets, respectively. CONCLUSIONS: ngLOC is a generic method that can be trained by data from a variety of species or classes for predicting protein subcellular localization. The standalone software is freely available for academic use under GNU GPL, and the ngLOC web server is also accessible at http://ngloc.unmc.edu.     

pLoc_bal-mGpos: Predict subcellular localization of Gram-positive bacterial proteins by quasi-balancing training dataset and PseAAC

Knowledge of protein subcellular localization is vitally important for both basic research and drug development. With the avalanche of protein sequences emerging in the post-genomic age, it is highly desired to develop computational tools for timely and effectively identifying their subcellular localization purely based on the sequence information alone. Recently, a predictor called “pLoc-mGpos” was developed for identifying the subcellular localization of Gram-positive bacterial proteins. Its performance is overwhelmingly better than that of the other predictors for the same purpose, particularly in dealing with multi-label systems in which some proteins, called “multiplex proteins”, may simultaneously occur in two or more subcellular locations. Although it is indeed a very powerful predictor, more efforts are definitely needed to further improve it. This is because pLoc-mGpos was trained by an extremely skewed dataset in which some subset (subcellular location) was over 11 times the size of the other subsets. Accordingly, it cannot avoid the bias consequence caused by such an uneven training dataset. To alleviate such bias consequence, we have developed a new and bias-reducing predictor called pLoc_bal-mGpos by quasi-balancing the training dataset. Rigorous target jackknife tests on exactly the same experiment-confirmed dataset have indicated that the proposed new predictor is remarkably superior to pLoc-mGpos, the existing state-of-the-art predictor in identifying the subcellular localization of Gram-positive bacterial proteins. To maximize the convenience for most experimental scientists, a user-friendly web-server for the new predictor has been established at http://www.jci-bioinfo.cn/pLoc_bal-mGpos/, by which users can easily get their desired results without the need to go through the detailed mathematics.     

MetaLocGramN: A meta-predictor of protein subcellular localization for Gram-negative bacteria

Subcellular localization is a key functional characteristic of proteins. It is determined by signals encoded in the protein sequence. The experimental determination of subcellular localization is laborious. Thus, a number of computational methods have been developed to predict the protein location from sequence. However predictions made by different methods often disagree with each other and it is not always clear which algorithm performs best for the given cellular compartment. We benchmarked primary subcellular localization predictors for proteins from Gram-negative bacteria, PSORTb3, PSLpred, CELLO, and SOSUI-GramN, on a common dataset that included 1056 proteins. We found that PSORTb3 performs best on the average, but is outperformed by other methods in predictions of extracellular proteins. This motivated us to develop a meta-predictor, which combines the primary methods by using the logistic regression models, to take advantage of their combined strengths, and to eliminate their individual weaknesses. MetaLocGramN runs the primary methods, and based on their output classifies protein sequences into one of five major localizations of the Gram-negative bacterial cell: cytoplasm, plasma membrane, periplasm, outer membrane, and extracellular space. MetaLocGramN achieves the average Matthews correlation coefficient of 0.806, i.e. 12% better than the best individual primary method. MetaLocGramN is a meta-predictor specialized in predicting subcellular localization for proteins from Gram-negative bacteria. According to our benchmark, it performs better than all other tools run independently. MetaLocGramN is a web and SOAP server available for free use by all academic users at the URL http://iimcb.genesilico.pl/MetaLocGramN. This article is part of a Special Issue entitled: Computational Methods for Protein Interaction and Structural Prediction.     

Manifestations of multicellularity: Dictyostelium reports in

The recent release of the Dictyostelium genome sequence is important because Dictyostelium has become a much-favoured model system for cell and developmental biologists. The sequence has revealed a remarkably high total number of approximately 12 500 genes, only a thousand fewer than are encoded by Drosophila. Previous protein-sequence comparisons suggested that Dictyostelium is evolutionarily closer to animals and fungi than to plants, and the global protein sequence comparison, now made possible by the genome sequence, confirms this. This review focuses on several classes of proteins that are shared by Dictyostelium and animals: a highly sophisticated array of microfilament components, a large family of G-protein-coupled receptors and a diverse set of SH2 domain-containing proteins. The presence of these proteins strengthens the case for a relatively close relationship with animals and extends the range of problems that can be addressed using Dictyostelium as a model organism.     

Predicting subcellular localization of proteins for Gram-negative bacteria by support vector machines based on n-peptide compositions

Gram-negative bacteria have five major subcellular localization sites: the cytoplasm, the periplasm, the inner membrane, the outer membrane, and the extracellular space. The subcellular location of a protein can provide valuable information about its function. With the rapid increase of sequenced genomic data, the need for an automated and accurate tool to predict subcellular localization becomes increasingly important. We present an approach to predict subcellular localization for Gram-negative bacteria. This method uses the support vector machines trained by multiple feature vectors based on n-peptide compositions. For a standard data set comprising 1443 proteins, the overall prediction accuracy reaches 89%, which, to the best of our knowledge, is the highest prediction rate ever reported. Our prediction is 14% higher than that of the recently developed multimodular PSORT-B. Because of its simplicity, this approach can be easily extended to other organisms and should be a useful tool for the high-throughput and large-scale analysis of proteomic and genomic data.     

Genome-wide identification of Arabidopsis coiled-coil proteins and establishment of the ARABI-COIL database

Increasing evidence demonstrates the importance of long coiled-coil proteins for the spatial organization of cellular processes. Although several protein classes with long coiled-coil domains have been studied in animals and yeast, our knowledge about plant long coiled-coil proteins is very limited. The repeat nature of the coiled-coil sequence motif often prevents the simple identification of homologs of animal coiled-coil proteins by generic sequence similarity searches. As a consequence, counterparts of many animal proteins with long coiled-coil domains, like lamins, golgins, or microtubule organization center components, have not been identified yet in plants. Here, all Arabidopsis proteins predicted to contain long stretches of coiled-coil domains were identified by applying the algorithm MultiCoil to a genome-wide screen. A searchable protein database, ARABI-COIL (http://www.coiled-coil.org/arabidopsis), was established that integrates information on number, size, and position of predicted coiled-coil domains with subcellular localization signals, transmembrane domains, and available functional annotations. ARABI-COIL serves as a tool to sort and browse Arabidopsis long coiled-coil proteins to facilitate the identification and selection of candidate proteins of potential interest for specific research areas. Using the database, candidate proteins were identified for Arabidopsis membrane-bound, nuclear, and organellar long coiled-coil proteins.     

Identification and characterization of a novel gene family YPEL in a wide spectrum of eukaryotic species

During comprehensive sequence analysis of human chromosome 22, we identified a novel gene family consisting of five members (YPEL1 through YPEL5) which has high homology with Drosophila yippee gene. We cloned and sequenced cDNAs for all five genes and determined their exon/intron organization. These YPEL genes showed high homology (43.8-96.6%) at amino acid sequence level among them. Mouse counterparts (Ypel1 through Ypel5) were also identified in the syntenic region of mouse chromosomes and their cDNAs were cloned and sequenced. Each of five pairs of human/mouse orthologs revealed extremely high homology. Thus, we named these genes as members of YPEL gene family. We searched YPEL family genes from the public databases, and found 100 genes from 68 species including animals, plants and fungi. Amino acid sequences of these 100 YPEL proteins were extremely similar and a consensus sequence of C-X(2)-C-X(19)-G-X(3)-L-X(5)-N-X(13)-G-X(8)-C-X(2)-C-X(4)-GWXY-X(10)-K-X(6)-E was established for all the YPEL family proteins without exception. Interestingly, the indirect immunofluorescent staining indicated that YPEL1-4 proteins are localized to the centrosome and nucleolus during interphase and at several dot-like structures around the mitotic apparatus during mitotic phase of COS-7 cells. YPEL5 protein is localized to the centrosome and nucleus during interphase and at the mitotic spindle during mitosis of the same cell line. Thus, the YPEL family proteins were found in essentially all the eukaryotes and hence they must play important roles in the maintenance of life. The subcellular localization of YPEL proteins in association with centrosome or mitotic spindle suggests a novel function involved in the cell division.     

PLPD: reliable protein localization prediction from imbalanced and overlapped datasets

Subcellular localization is one of the key functional characteristics of proteins. An automatic and efficient prediction method for the protein subcellular localization is highly required owing to the need for large-scale genome analysis. From a machine learning point of view, a dataset of protein localization has several characteristics: the dataset has too many classes (there are more than 10 localizations in a cell), it is a multi-label dataset (a protein may occur in several different subcellular locations), and it is too imbalanced (the number of proteins in each localization is remarkably different). Even though many previous works have been done for the prediction of protein subcellular localization, none of them tackles effectively these characteristics at the same time. Thus, a new computational method for protein localization is eventually needed for more reliable outcomes. To address the issue, we present a protein localization predictor based on D-SVDD (PLPD) for the prediction of protein localization, which can find the likelihood of a specific localization of a protein more easily and more correctly. Moreover, we introduce three measurements for the more precise evaluation of a protein localization predictor. As the results of various datasets which are made from the experiments of Huh et al. (2003), the proposed PLPD method represents a different approach that might play a complimentary role to the existing methods, such as Nearest Neighbor method and discriminate covariant method. Finally, after finding a good boundary for each localization using the 5184 classified proteins as training data, we predicted 138 proteins whose subcellular localizations could not be clearly observed by the experiments of Huh et al. (2003).     

Analysis of histone acetyltransferase and histone deacetylase families of Arabidopsis thaliana suggests functional diversification of chromatin modification among multicellular eukaryotes

Sequence similarity and profile searching tools were used to analyze the genome sequences of Arabidopsis thaliana, Saccharomyces cerevisiae, Schizosaccharomyces pombe, Caenorhabditis elegans and Drosophila melanogaster for genes encoding three families of histone deacetylase (HDAC) proteins and three families of histone acetyltransferase (HAT) proteins. Plants, animals and fungi were found to have a single member of each of three subfamilies of the GNAT family of HATs, suggesting conservation of these functions. However, major differences were found with respect to sizes of gene families and multi-domain protein structures within other families of HATs and HDACs, indicating substantial evolutionary diversification. Phylogenetic analysis identified a new class of HDACs within the RPD3/HDA1 family that is represented only in plants and animals. A similar analysis of the plant-specific HD2 family of HDACs suggests a duplication event early in dicot evolution, followed by further diversification in the lineage leading to Arabidopsis. Of three major classes of SIR2-type HDACs that are found in animals, fungi have representatives only in one class, whereas plants have representatives only in the other two. Plants possess five CREB-binding protein (CBP)-type HATs compared with one to two in animals and none in fungi. Domain and phylogenetic analyses of the CBP family proteins showed that this family has evolved three distinct types of CBPs in plants. The domain architecture of CBP and TAF(II)250 families of HATs show significant differences between plants and animals, most notably with respect to bromodomain occurrence and their number. Bromodomain-containing proteins in Arabidopsis differ strikingly from animal bromodomain proteins with respect to the numbers of bromodomains and the other types of domains that are present. The substantial diversification of HATs and HDACs that has occurred since the divergence of plants, animals and fungi suggests a surprising degree of evolutionary plasticity and functional diversification in these core chromatin components.     

Prediction of protein subcellular localization by weighted gene ontology terms

We develop a new weighting approach of gene ontology (GO) terms for predicting protein subcellular localization. The weights of individual GO terms, corresponding to their contribution to the prediction algorithm, are determined by the term-weighting methods used in text categorization. We evaluate several term-weighting methods, which are based on inverse document frequency, information gain, gain ratio, odds ratio, and chi-square and its variants. Additionally, we propose a new term-weighting method based on the logarithmic transformation of chi-square. The proposed term-weighting method performs better than other term-weighting methods, and also outperforms state-of-the-art subcellular prediction methods. Our proposed method achieves 98.1%, 99.3%, 98.1%, 98.1%, and 95.9% overall accuracies for the animal BaCelLo independent dataset (IDS), fungal BaCelLo IDS, animal Höglund IDS, fungal Höglund IDS, and PLOC dataset, respectively. Furthermore, the close correlation between high-weighted GO terms and subcellular localizations suggests that our proposed method appropriately weights GO terms according to their relevance to the localizations.     

A multi-label classifier for predicting the subcellular localization of gram-negative bacterial proteins with both single and multiple sites

Prediction of protein subcellular localization is a challenging problem, particularly when the system concerned contains both singleplex and multiplex proteins. In this paper, by introducing the "multi-label scale" and hybridizing the information of gene ontology with the sequential evolution information, a novel predictor called iLoc-Gneg is developed for predicting the subcellular localization of gram-positive bacterial proteins with both single-location and multiple-location sites. For facilitating comparison, the same stringent benchmark dataset used to estimate the accuracy of Gneg-mPLoc was adopted to demonstrate the power of iLoc-Gneg. The dataset contains 1,392 gram-negative bacterial proteins classified into the following eight locations: (1) cytoplasm, (2) extracellular, (3) fimbrium, (4) flagellum, (5) inner membrane, (6) nucleoid, (7) outer membrane, and (8) periplasm. Of the 1,392 proteins, 1,328 are each with only one subcellular location and the other 64 are each with two subcellular locations, but none of the proteins included has pairwise sequence identity to any other in a same subset (subcellular location). It was observed that the overall success rate by jackknife test on such a stringent benchmark dataset by iLoc-Gneg was over 91%, which is about 6% higher than that by Gneg-mPLoc. As a user-friendly web-server, iLoc-Gneg is freely accessible to the public at http://icpr.jci.edu.cn/bioinfo/iLoc-Gneg. Meanwhile, a step-by-step guide is provided on how to use the web-server to get the desired results. Furthermore, for the user's convenience, the iLoc-Gneg web-server also has the function to accept the batch job submission, which is not available in the existing version of Gneg-mPLoc web-server. It is anticipated that iLoc-Gneg may become a useful high throughput tool for Molecular Cell Biology, Proteomics, System Biology, and Drug Development.     

Protein subcellular localization based on PSI-BLAST and machine learning

Subcellular location is an important functional annotation of proteins. An automatic, reliable and efficient prediction system for protein subcellular localization is necessary for large-scale genome analysis. This paper describes a protein subcellular localization method which extracts features from protein profiles rather than from amino acid sequences. The protein profile represents a protein family, discards part of the sequence information that is not conserved throughout the family and therefore is more sensitive than the amino acid sequence. The amino acid compositions of whole profile and the N-terminus of the profile are extracted, respectively, to train and test the probabilistic neural network classifiers. On two benchmark datasets, the overall accuracies of the proposed method reach 89.1% and 68.9%, respectively. The prediction results show that the proposed method perform better than those methods based on amino acid sequences. The prediction results of the proposed method are also compared with Subloc on two redundance-reduced datasets.     

基于序列拓扑和二阶隐马尔可夫模型的跨膜蛋白亚细胞定位预测

现有蛋白质亚细胞定位方法针对水溶性蛋白质而设计,对跨膜蛋白并不适用。而专门的跨膜拓扑预测器,又不是为亚细胞定位而设计的。文章改进了跨膜拓扑预测器TMPHMMLoc的模型结构,设计了一个新的二阶隐马尔可夫模型;采用推广到二阶模型的Baum-Welch算法估计模型参数,并把将各个亚细胞位置建立的模型整合为一个预测器。数据集上测试结果表明,此方法性能显著优于针对可溶性蛋白设计的支持向量机方法和模糊k最邻近方法,也优于TMPHMMLoc中提出的隐马尔可夫模型方法,是一个有效的跨膜蛋白亚细胞定位预测方法。     

用离散增量结合支持向量机方法预测蛋白质亚细胞定位

对未知蛋白的功能注释是蛋白质组学的主要目标。一个关键的注释是蛋白质亚细胞定位的预测。本文应用离散增量结合支持向量机(ID_SVM)的方法,对阳性革兰氏细菌蛋白的5类亚细胞定位点进行预测。在独立检验下,其总体预测成功率为89.66%。结果发现ID_SVM算法对预测的成功率有很大改进。     

Addressing protein localization within the nucleus

Bridging the gap between the number of gene sequences in databases and the number of gene products that have been functionally characterized in any way is a major challenge for biology. A key characteristic of proteins, which can begin to elucidate their possible functions, is their subcellular location. A number of experimental approaches can reveal the subcellular localization of proteins in mammalian cells. However, genome databases now contain predicted sequences for a large number of potentially novel proteins that have yet to be studied in any way, let alone have their subcellular localization determined. Here we ask whether using bioinformatics tools to analyse the sequence of proteins whose subnuclear localizations have been determined can reveal characteristics or signatures that might allow us to predict localization for novel protein sequences.