Identification of the subcellular localization of mycobacterial proteins using localization motifs

Mycobacterium, the most common disease-causing genus, infects billions of people and is notoriously difficult to treat. Understanding the subcellular localization of mycobacterial proteins can provide essential clues for protein function and drug discovery. In this article, we present a novel approach that focuses on local sequence information to identify localization motifs that are generated by a merging algorithm and are selected based on a binomially distributed model. These localization motifs are employed as features for identifying the subcellular localization of mycobacterial proteins. Our approach provides more accurate results than previous methods and was tested on an independent dataset recently obtained from an experimental study to provide a first and reasonably accurate prediction of subcellular localization. Our approach can also be used for large-scale prediction of new protein entries in the UniportKB database and of protein sequences obtained experimentally. In addition, our approach identified many local motifs involved with the subcellular localization that also interact with the environment. Thus, our method may have widespread applications both in the study of the functions of mycobacterial proteins and in the search for a potential vaccine target for designing drugs.     

Esub8: A novel tool to predict protein subcellular localizations in eukaryotic organisms

Background

Subcellular localization of a new protein sequence is very important and fruitful for understanding its function. As the number of new genomes has dramatically increased over recent years, a reliable and efficient system to predict protein subcellular location is urgently needed.

Results

Esub8 was developed to predict protein subcellular localizations for eukaryotic proteins based on amino acid composition. In this research, the proteins are classified into the following eight groups: chloroplast, cytoplasm, extracellular, Golgi apparatus, lysosome, mitochondria, nucleus and peroxisome. We know subcellular localization is a typical classification problem; consequently, a one-against-one (1-v-1) multi-class support vector machine was introduced to construct the classifier. Unlike previous methods, ours considers the order information of protein sequences by a different method. Our method is tested in three subcellular localization predictions for prokaryotic proteins and four subcellular localization predictions for eukaryotic proteins on Reinhardt's dataset. The results are then compared to several other methods. The total prediction accuracies of two tests are both 100% by a self-consistency test, and are 92.9% and 84.14% by the jackknife test, respectively. Esub8 also provides excellent results: the total prediction accuracies are 100% by a self-consistency test and 87% by the jackknife test.

Conclusions

Our method represents a different approach for predicting protein subcellular localization and achieved a satisfactory result; furthermore, we believe Esub8 will be a useful tool for predicting protein subcellular localizations in eukaryotic organisms.

     

An FPT approach for predicting protein localization from yeast genomic data

Accurately predicting the localization of proteins is of paramount importance in the quest to determine their respective functions within the cellular compartment. Because of the continuous and rapid progress in the fields of genomics and proteomics, more data are available now than ever before. Coincidentally, data mining methods been developed and refined in order to handle this experimental windfall, thus allowing the scientific community to quantitatively address long-standing questions such as that of protein localization. Here, we develop a frequent pattern tree (FPT) approach to generate a minimum set of rules (mFPT) for predicting protein localization. We acquire a series of rules according to the features of yeast genomic data. The mFPT prediction accuracy is benchmarked against other commonly used methods such as Bayesian networks and logistic regression under various statistical measures. Our results show that mFPT gave better performance than other approaches in predicting protein localization. Meanwhile, setting 0.65 as the minimum hit-rate, we obtained 138 proteins that mFPT predicted differently than the simple naive bayesian method (SNB). In our analysis of these 138 proteins, we present novel predictions for the location for 17 proteins, which currently do not have any defined localization. These predictions can serve as putative annotations and should provide preliminary clues for experimentalists. We also compared our predictions against the eukaryotic subcellular localization database and related predictions by others on protein localization. Our method is quite generalized and can thus be applied to discover the underlying rules for protein-protein interactions, genomic interactions, and structure-function relationships, as well as those of other fields of research.     

人CDC73基因编码蛋白的理化性质及功能研究

采用生物信息学方法对人CDC73 (cell division cycle 73)基因编码蛋白的理化性质、亲疏水性、跨膜区域、信号肽区域、二级结构、三级结构、蛋白质之间的相互作用、亚细胞定位进行预测分析。使用多种分析软件对人CDC73基因编码蛋白进行预测分析。研究可知,人CDC73基因属于抑癌基因,该基因编码一个由531个氨基酸组成的肿瘤抑制因子Parafibromin,其等电点为9.63,半衰期为30 h且在哺乳动物中高度保守;二级结构预测发现13个α螺旋和10个β折叠片层,三级结构预测结果的可靠性达69.01%,亚细胞定位主要分布于细胞质及细胞核。由本研究可知,人CDC73基因编码蛋白是一个存在核定位序列的不稳定亲水蛋白,在细胞内广泛分布并参与多种生命活动过程,且能够抑制肿瘤的产生。对人CDC73基因编码蛋白结构和功能的预测分析,可为其进一步的研究提供一定的理论依据,也为相关疾病的诊治提供新的思路。     

Hidden localization motifs: naturally occurring peroxisomal targeting signals in non-peroxisomal proteins

Background

Can sequence segments coding for subcellular targeting or for posttranslational modifications occur in proteins that are not substrates in either of these processes? Although considerable effort has been invested in achieving low false-positive prediction rates, even accurate sequence-analysis tools for the recognition of these motifs generate a small but noticeable number of protein hits that lack the appropriate biological context but cannot be rationalized as false positives.

Results

We show that the carboxyl termini of a set of definitely non-peroxisomal proteins with predicted peroxisomal targeting signals interact with the peroxisomal matrix protein receptor peroxin 5 (PEX5) in a yeast two-hybrid test. Moreover, we show that examples of these proteins - chicken lysozyme, human tyrosinase and the yeast mitochondrial ribosomal protein L2 (encoded by MRP7) - are imported into peroxisomes in vivo if their original sorting signals are disguised. We also show that even prokaryotic proteins can contain peroxisomal targeting sequences.

Conclusions

Thus, functional localization signals can evolve in unrelated protein sequences as a result of neutral mutations, and subcellular targeting is hierarchically organized, with signal accessibility playing a decisive role. The occurrence of silent functional motifs in unrelated proteins is important for the development of sequence-based function prediction tools and the interpretation of their results. Silent functional signals have the potential to acquire importance in future evolutionary scenarios and in pathological conditions.     

Characterization of human proteins with different subcellular localizations by topological and biological properties

Knowing the protein localization can provide valuable information resource for elucidating protein function. In recent years, with the advances of human genomics and proteomics, it is possible to characterize human proteins that are located in different subcellular localizations. In this study, we used the topological properties and biological properties to characterize human proteins with six subcellular localizations. Almost all of these properties were found to be significantly different among six protein categories. Network topology analysis indicated that several significant topological properties, including the degree and k-core, were higher for the mitochondrial proteins. Biological property analysis showed that the nuclear proteins appeared to be correlated with important biological function. We hope these findings may provide some important help for comprehensive understanding the biological function of proteins, and prediction of protein subcellular localizations in human.     

FGsub: <Emphasis Type="Italic">Fusarium graminearum</Emphasis> protein subcellular localizations predicted from primary structures

Background

The fungal pathogen Fusarium graminearum (telomorph Gibberella zeae) is the causal agent of several destructive crop diseases, where a set of genes usually work in concert to cause diseases to crops. To function appropriately, the F. graminearum proteins inside one cell should be assigned to different compartments, i.e. subcellular localizations. Therefore, the subcellular localizations of F. graminearum proteins can provide insights into protein functions and pathogenic mechanisms of this destructive pathogen fungus. Unfortunately, there are no subcellular localization information for F. graminearum proteins available now. Computational approaches provide an alternative way to predicting F. graminearum protein subcellular localizations due to the expensive and time-consuming biological experiments in lab.

Results

In this paper, we developed a novel predictor, namely FGsub, to predict F. graminearum protein subcellular localizations from the primary structures. First, a non-redundant fungi data set with subcellular localization annotation is collected from UniProtKB database and used as training set, where the subcellular locations are classified into 10 groups. Subsequently, Support Vector Machine (SVM) is trained on the training set and used to predict F. graminearum protein subcellular localizations for those proteins that do not have significant sequence similarity to those in training set. The performance of SVMs on training set with 10-fold cross-validation demonstrates the efficiency and effectiveness of the proposed method. In addition, for F. graminearum proteins that have significant sequence similarity to those in training set, BLAST is utilized to transfer annotations of homologous proteins to uncharacterized F. graminearum proteins so that the F. graminearum proteins are annotated more comprehensively.

Conclusions

In this work, we present FGsub to predict F. graminearum protein subcellular localizations in a comprehensive manner. We make four fold contributions to this filed. First, we present a new algorithm to cope with imbalance problem that arises in protein subcellular localization prediction, which can solve imbalance problem and avoid false positive results. Second, we design an ensemble classifier which employs feature selection to further improve prediction accuracy. Third, we use BLAST to complement machine learning based methods, which enlarges our prediction coverage. Last and most important, we predict the subcellular localizations of 12786 F. graminearum proteins, which provide insights into protein functions and pathogenic mechanisms of this destructive pathogen fungus.

     

Approaches to defining dual-targeted proteins in Arabidopsis

A variety of approaches were used to predict dual-targeted proteins in Arabidopsis thaliana . These predictions were experimentally tested using GFP fusions. Twelve new dual-targeted proteins were identified: five that were dual-targeted to mitochondria and plastids, six that were dual-targeted to mitochondria and peroxisomes, and one that was dual-targeted to mitochondria and the nucleus. Two methods to predict dual-targeted proteins had a high success rate: (1) combining the AraPerox database with a variety of subcellular prediction programs to identify mitochondrial- and peroxisomal-targeted proteins, and (2) using a variety of prediction programs on a biochemical pathway or process known to contain at least one dual-targeted protein. Several technical parameters need to be taken into account before assigning subcellular localization using GFP fusion proteins. The position of GFP with respect to the tagged polypeptide, the tissue or cells used to detect subcellular localization, and the portion of a candidate protein fused to GFP are all relevant to the expression and targeting of a fusion protein. Testing all gene models for a chromosomal locus is required if more than one model exists.     

Multi-label multi-kernel transfer learning for human protein subcellular localization

Recent years have witnessed much progress in computational modelling for protein subcellular localization. However, the existing sequence-based predictive models demonstrate moderate or unsatisfactory performance, and the gene ontology (GO) based models may take the risk of performance overestimation for novel proteins. Furthermore, many human proteins have multiple subcellular locations, which renders the computational modelling more complicated. Up to the present, there are far few researches specialized for predicting the subcellular localization of human proteins that may reside in multiple cellular compartments. In this paper, we propose a multi-label multi-kernel transfer learning model for human protein subcellular localization (MLMK-TLM). MLMK-TLM proposes a multi-label confusion matrix, formally formulates three multi-labelling performance measures and adapts one-against-all multi-class probabilistic outputs to multi-label learning scenario, based on which to further extends our published work GO-TLM (gene ontology based transfer learning model for protein subcellular localization) and MK-TLM (multi-kernel transfer learning based on Chou's PseAAC formulation for protein submitochondria localization) for multiplex human protein subcellular localization. With the advantages of proper homolog knowledge transfer, comprehensive survey of model performance for novel protein and multi-labelling capability, MLMK-TLM will gain more practical applicability. The experiments on human protein benchmark dataset show that MLMK-TLM significantly outperforms the baseline model and demonstrates good multi-labelling ability for novel human proteins. Some findings (predictions) are validated by the latest Swiss-Prot database. The software can be freely downloaded at http://soft.synu.edu.cn/upload/msy.rar.     

Iterative orthology prediction uncovers new mitochondrial proteins and identifies C12orf62 as the human ortholog of COX14, a protein involved in the assembly of cytochrome c oxidase

Background

Orthology is a central tenet of comparative genomics and ortholog identification is instrumental to protein function prediction. Major advances have been made to determine orthology relations among a set of homologous proteins. However, they depend on the comparison of individual sequences and do not take into account divergent orthologs.

Results

We have developed an iterative orthology prediction method, Ortho-Profile, that uses reciprocal best hits at the level of sequence profiles to infer orthology. It increases ortholog detection by 20% compared to sequence-to-sequence comparisons. Ortho-Profile predicts 598 human orthologs of mitochondrial proteins from Saccharomyces cerevisiae and Schizosaccharomyces pombe with 94% accuracy. Of these, 181 were not known to localize to mitochondria in mammals. Among the predictions of the Ortho-Profile method are 11 human cytochrome c oxidase (COX) assembly proteins that are implicated in mitochondrial function and disease. Their co-expression patterns, experimentally verified subcellular localization, and co-purification with human COX-associated proteins support these predictions. For the human gene C12orf62, the ortholog of S. cerevisiae COX14, we specifically confirm its role in negative regulation of the translation of cytochrome c oxidase.

Conclusions

Divergent homologs can often only be detected by comparing sequence profiles and profile-based hidden Markov models. The Ortho-Profile method takes advantage of these techniques in the quest for orthologs.     

High-throughput protein localization in Arabidopsis using Agrobacterium-mediated transient expression of GFP-ORF fusions

We describe a streamlined and systematic method for cloning green fluorescent protein (GFP)-open reading frame (ORF) fusions and assessing their subcellular localization in Arabidopsis thaliana cells. The sequencing of the Arabidopsis genome has made it feasible to undertake genome-based approaches to determine the function of each protein and define its subcellular localization. This is an essential step towards full functional analysis. The approach described here allows the economical handling of hundreds of expressed plant proteins in a timely fashion. We have integrated recombinational cloning of full-length trimmed ORF clones (available from the SSP consortium) with high-efficiency transient transformation of Arabidopsis cell cultures by a hypervirulent strain of Agrobacterium. To demonstrate its utility, we have used a selection of trimmed ORFs, representing a variety of key cellular processes and have defined the localization patterns of 155 fusion proteins. These patterns have been classified into five main categories, including cytoplasmic, nuclear, nucleolar, organellar and endomembrane compartments. Several genes annotated in GenBank as unknown have been ascribed a protein localization pattern. We also demonstrate the application of flow cytometry to estimate the transformation efficiency and cell cycle phase of the GFP-positive cells. This approach can be extended to functional studies, including the precise cellular localization and the prediction of the role of unknown proteins, the confirmation of bioinformatic predictions and proteomic experiments, such as the determination of protein interactions in vivo, and therefore has numerous applications in the post-genomic analysis of protein function.     

Application of the accurate mass and time tag approach to the proteome analysis of sub-cellular fractions obtained from Rhodobacter sphaeroides 2.4.1. Aerobic and photosynthetic cell cultures

The high-throughput accurate mass and time (AMT) tag proteomic approach was utilized to characterize the proteomes for cytoplasm, cytoplasmic membrane, periplasm, and outer membrane fractions from aerobic and photosynthetic cultures of the gram-nagtive bacterium Rhodobacter sphaeroides 2.4.1. In addition, we analyzed the proteins within purified chromatophore fractions that house the photosynthetic apparatus from photosynthetically grown cells. In total, 8,300 peptides were identified with high confidence from at least one subcellular fraction from either cell culture. These peptides were derived from 1,514 genes or 35% percent of proteins predicted to be encoded by the genome. A significant number of these proteins were detected within a single subcellular fraction and their localization was compared to in silico predictions. However, the majority of proteins were observed in multiple subcellular fractions, and the most likely subcellular localization for these proteins was investigated using a Z-score analysis of estimated protein abundance along with clustering techniques. Good (81%) agreement was observed between the experimental results and in silico predictions. The AMT tag approach provides localization evidence for those proteins that have no predicted localization information, those annotated as putative proteins, and/or for those proteins annotated as hypothetical and conserved hypothetical.     

Refining protein subcellular localization

The study of protein subcellular localization is important to elucidate protein function. Even in well-studied organisms such as yeast, experimental methods have not been able to provide a full coverage of localization. The development of bioinformatic predictors of localization can bridge this gap. We have created a Bayesian network predictor called PSLT2 that considers diverse protein characteristics, including the combinatorial presence of InterPro motifs and protein interaction data. We compared the localization predictions of PSLT2 to high-throughput experimental localization datasets. Disagreements between these methods generally involve proteins that transit through or reside in the secretory pathway. We used our multi-compartmental predictions to refine the localization annotations of yeast proteins primarily by distinguishing between soluble lumenal proteins and soluble proteins peripherally associated with organelles. To our knowledge, this is the first tool to provide this functionality. We used these sub-compartmental predictions to characterize cellular processes on an organellar scale. The integration of diverse protein characteristics and protein interaction data in an appropriate setting can lead to high-quality detailed localization annotations for whole proteomes. This type of resource is instrumental in developing models of whole organelles that provide insight into the extent of interaction and communication between organelles and help define organellar functionality.     

基于茶蛋白质组学数据分析植物亚细胞定位预测软件的应用

以500个茶（Camellia sinensis（L.）O.Ktze.）叶片的蛋白质作为数据集,比较TargetP、WoLF PSORT、LocTree和Plant-mPLoc 4种软件预测亚细胞定位的可信度和灵敏度。结果显示,4种软件预测可信度均高于80%,依次排序为TargetP > LocTree > WoLF PSORT > Plant-mPLoc。其中,LocTree对细胞质蛋白和分泌蛋白检测灵敏度最高,但对叶绿体蛋白灵敏度最低;Plant-mPLoc检测核蛋白最灵敏,但对细胞质蛋白最不敏感;TargetP检测叶绿体蛋白最灵敏,但仅能区分3个亚细胞器官;WoLF PSORT对分泌蛋白检测灵敏度最低,但对其他蛋白均较灵敏。基于上述结果,该研究针对4种软件提出了合理的使用建议。     

Efficient and Interpretable Prediction of Protein Functional Classes by Correspondence Analysis and Compact Set Relations

Predicting protein functional classes such as localization sites and modifications plays a crucial role in function annotation. Given a tremendous amount of sequence data yielded from high-throughput sequencing experiments, the need of efficient and interpretable prediction strategies has been rapidly amplified. Our previous approach for subcellular localization prediction, PSLDoc, archives high overall accuracy for Gram-negative bacteria. However, PSLDoc is computational intensive due to incorporation of homology extension in feature extraction and probabilistic latent semantic analysis in feature reduction. Besides, prediction results generated by support vector machines are accurate but generally difficult to interpret. In this work, we incorporate three new techniques to improve efficiency and interpretability. First, homology extension is performed against a compact non-redundant database using a fast search model to reduce running time. Second, correspondence analysis (CA) is incorporated as an efficient feature reduction to generate a clear visual separation of different protein classes. Finally, functional classes are predicted by a combination of accurate compact set (CS) relation and interpretable one-nearest neighbor (1-NN) algorithm. Besides localization data sets, we also apply a human protein kinase set to validate generality of our proposed method. Experiment results demonstrate that our method make accurate prediction in a more efficient and interpretable manner. First, homology extension using a fast search on a compact database can greatly accelerate traditional running time up to twenty-five times faster without sacrificing prediction performance. This suggests that computational costs of many other predictors that also incorporate homology information can be largely reduced. In addition, CA can not only efficiently identify discriminative features but also provide a clear visualization of different functional classes. Moreover, predictions based on CS achieve 100% precision. When combined with 1-NN on unpredicted targets by CS, our method attains slightly better or comparable performance compared with the state-of-the-art systems.     

Mapping the subcellular protein distribution in three human cell lines

The subcellular locations of proteins are closely related to their function and constitute an essential aspect for understanding the complex machinery of living cells. A systematic effort has been initiated to map the protein distribution in three functionally different cell lines with the aim to provide a subcellular localization index for at least one representative protein from all human protein-encoding genes. Here, we present the results of more than 3500 proteins mapped to 16 subcellular compartments. The results indicate a ubiquitous protein expression with a majority of the proteins found in all three cell lines and a large portion localized to two or more compartments. The inter-relationships between the subcellular compartments are visualized in a protein-compartment network based on all detected proteins. Hierarchical clustering was performed to determine how closely related the organelles are in terms of protein constituents and compare the proteins detected in each cell type. Our results show distinct organelle proteomes, well conserved across the cell types, and demonstrate that biochemically similar organelles are grouped together.     

Predicting subcellular localization with AdaBoost Learner

Protein subcellular localization, which tells where a protein resides in a cell, is an important characteristic of a protein, and relates closely to the function of proteins. The prediction of their subcellular localization plays an important role in the prediction of protein function, genome annotation and drug design. Therefore, it is an important and challenging role to predict subcellular localization using bio-informatics approach. In this paper, a robust predictor, AdaBoost Learner is introduced to predict protein subcellular localization based on its amino acid composition. Jackknife cross-validation and independent dataset test were used to demonstrate that Adaboost is a robust and efficient model in predicting protein subcellular localization. As a result, the correct prediction rates were 74.98% and 80.12% for the Jackknife test and independent dataset test respectively, which are higher than using other existing predictors. An online server for predicting subcellular localization of proteins based on AdaBoost classifier was available on http://chemdata.shu. edu.cn/sl12.     

The sequences of MinE responsible for its subcellular localization analyzed by competitive binding method in <Emphasis Type="Italic">Escherichia coli</Emphasis>

The subcellular localization of a protein is important for its proper function. Escherichia coli MinE is a small protein with clear subcellular localization, which provides a good model to study protein localization mechanism. In the present study, a series of recombinant minEs truncated in one end or in the middle regions, fused with egfp, was constructed, and these recombinant proteins could compete to function with the chromosomal MinE. Our results showed that the sequences related to the subcellular localization of MinE span several functional domains, demonstrating that MinE positioning in cells depends on multiple factors. The eGFP fusions with some truncated MinE from N-terminal resulted in different cell phenotypes and localization features, implying that these fusions can interfere chromosomal MinE’s function, similar to MinE^36–88 phenotype in the previous report. The amino acid in the region (32–48) is sensitive to change MinE conformation and influence its dimerization. Some truncated protein structure could be unstable. Thus, the MinE localization is prerequisite for its proper anti-MinCD function and some new features of MinE were demonstrated. This approach can be extended for subcellular localization research for other essential proteins.