Gene ontology based transfer learning for protein subcellular localization

A Review for Artificial Intelligence Based Protein Subcellular Localization

Proteins need to be located in appropriate spatiotemporal contexts to carry out their diverse biological functions. Mislocalized proteins may lead to a broad range of diseases, such as cancer and Alzheimer's disease. Knowing where a target protein resides within a cell will give insights into tailored drug design for a disease. As the gold validation standard, the conventional wet lab uses fluorescent microscopy imaging, immunoelectron microscopy, and fluorescent biomarker tags for protein subcellular location identification. However, the booming era of proteomics and high-throughput sequencing generates tons of newly discovered proteins, making protein subcellular localization by wet-lab experiments a mission impossible. To tackle this concern, in the past decades, artificial intelligence (AI) and machine learning (ML), especially deep learning methods, have made significant progress in this research area. In this article, we review the latest advances in AI-based method development in three typical types of approaches, including sequence-based, knowledge-based, and image-based methods. We also elaborately discuss existing challenges and future directions in AI-based method development in this research field.

Disclosing the locale of transmembrane proteins within cellular alcove by machine learning approach: systematic review and meta analysis

Protein subcellular localization is a promising research question in Proteomics and associated fields, including Biological Sciences, Biomedical Engineering, Computational Biology, Bioinformatics, Proteomics, Artificial Intelligence, and Biophysics. However, computational techniques are preferred to explore this attribute for a massive number of proteins. The byproduct of this conjunction yields diversified location identifiers of proteins. These protein subcellular localization identifiers are unique regarding the database used, organisms, Machine Learning Technique, and accuracy. Despite the availability of these identifiers, the majority of the work has been done on the subcellular localization of proteins and, less work has been done specifically on locations of transmembrane proteins. This systematic review accounts for computational techniques implemented on transmembrane protein localization. Moreover, a literature search on PubMed, Science Direct, and IEEE Databases disclosed no systematic review or meta-analysis on the cell's transmembrane protein locale. A Systematic review was formed under the guidelines of PRISMA by using Science Direct, PubMed, and IEEE Databases. Journal publications from 2000 to 2023 were taken into consideration and screened. This review has focused only on computational studies rather than experimental techniques. 1004 studies were reviewed and were categorized as relevant and non-relevant according to inclusion and exclusion criteria. All the screening was done through Endnote after importing citations. This systematic review characterizes the gap in targeting the locale of the transmembrane protein and will aid researchers in exploring its new horizons.Communicated by Ramaswamy H. Sarma.

Spectral Encoder to Extract the Features of Near-Infrared Spectra for Multivariate Calibration

An autoencoder architecture was adopted for near-infrared (NIR) spectral analysis by extracting the common features in the spectra. Three autoencoder-based networks with different purposes were constructed. First, a spectral encoder was established by training the network with a set of spectra as the input. The features of the spectra can be encoded by the nodes in the bottleneck layer, which in turn can be used to build a sparse and robust model. Second, taking the spectra of one instrument as the input and that of another instrument as the reference output, the common features in both spectra can be obtained in the bottleneck layer. Therefore, in the prediction step, the spectral features of the second can be predicted by taking the reverse of the decoder as the encoder. Furthermore, transfer learning was used to build the model for the spectra of more instruments by fine-tuning the trained network. NIR datasets of plant, wheat, and pharmaceutical tablets measured on multiple instruments were used to test the method. The multi-linear regression (MLR) model with the encoded features was found to have a similar or slightly better performance in prediction compared with the partial least-squares (PLS) model.

Transfer Learning for High-Dimensional Linear Regression: Prediction, Estimation and Minimax Optimality

This paper considers estimation and prediction of a high-dimensional linear regression in the setting of transfer learning where, in addition to observations from the target model, auxiliary samples from different but possibly related regression models are available. When the set of informative auxiliary studies is known, an estimator and a predictor are proposed and their optimality is established. The optimal rates of convergence for prediction and estimation are faster than the corresponding rates without using the auxiliary samples. This implies that knowledge from the informative auxiliary samples can be transferred to improve the learning performance of the target problem. When the set of informative auxiliary samples is unknown, we propose a data-driven procedure for transfer learning, called Trans-Lasso, and show its robustness to non-informative auxiliary samples and its efficiency in knowledge transfer. The proposed procedures are demonstrated in numerical studies and are applied to a dataset concerning the associations among gene expressions. It is shown that Trans-Lasso leads to improved performance in gene expression prediction in a target tissue by incorporating data from multiple different tissues as auxiliary samples.

A Review of Domain Adaptation without Target Labels

Domain adaptation has become a prominent problem setting in machine learning and related fields. This review asks the question: How can a classifier learn from a source domain and generalize to a target domain? We present a categorization of approaches, divided into, what we refer to as, sample-based, feature-based, and inference-based methods. Sample-based methods focus on weighting individual observations during training based on their importance to the target domain. Feature-based methods revolve around on mapping, projecting, and representing features such that a source classifier performs well on the target domain and inference-based methods incorporate adaptation into the parameter estimation procedure, for instance through constraints on the optimization procedure. Additionally, we review a number of conditions that allow for formulating bounds on the cross-domain generalization error. Our categorization highlights recurring ideas and raises questions important to further research.

Heterogeneous Domain Adaptation for IHC Classification of Breast Cancer Subtypes

Increasingly, multiple parallel omics datasets are collected from biological samples. Integrating these datasets for classification is an open area of research. Additionally, whilst multiple datasets may be available for the training samples, future samples may only be measured by a single technology requiring methods which do not rely on the presence of all datasets for sample prediction. This enables us to directly compare the protein and the gene profiles. New samples with just one set of measurements (e.g., just protein) can then be mapped to this latent common space where classification is performed. Using this approach, we achieved an improvement of up to 12 percent in accuracy when classifying samples based on their protein measurements compared with baseline methods which were trained on the protein data alone. We illustrate that the additional inclusion of the gene expression or protein expression in the training process enabled the separation between the classes to become clearer.

Transfer Learning for Molecular Cancer Classification Using Deep Neural Networks

The emergence of deep learning has impacted numerous machine learning based applications and research. The reason for its success lies in two main advantages: 1) it provides the ability to learn very complex non-linear relationships between features and 2) it allows one to leverage information from unlabeled data that does not belong to the problem being handled. This paper presents a transfer learning procedure for cancer classification, which uses feature selection and normalization techniques in conjunction with s sparse auto-encoders on gene expression data. While classifying any two tumor types, data of other tumor types were used in unsupervised manner to improve the feature representation. The performance of our algorithm was tested on 36 two-class benchmark datasets from the GEMLeR repository. On performing statistical tests, it is clearly ascertained that our algorithm statistically outperforms several generally used cancer classification approaches. The deep learning based molecular disease classification can be used to guide decisions made on the diagnosis and treatment of diseases, and therefore may have important applications in precision medicine.

ML-rRBF-ECOC: A Multi-Label Learning Classifier for Predicting Protein Subcellular Localization with Both Single and Multiple Sites

Background:

The subcellular localization of a protein is closely related with its functions and interactions. More and more evidences show that proteins may simultaneously exist at, or move between, two or more different subcellular localizations. Therefore, predicting protein subcellular localization is an important but challenging problem.

Observation:

Most of the existing methods for predicting protein subcellular localization assume that a protein locates at a single site. Although a few methods have been proposed to deal with proteins with multiple sites, correlations between subcellular localization are not efficiently taken into account. In this paper, we propose an integrated method for predicting protein subcellular localizations with both single site and multiple sites.

Methods:

Firstly, we extend the Multi-Label Radial Basis Function (ML-RBF) method to the regularized version, and augment the first layer of ML-RBF to take local correlations between subcellular localization into account. Secondly, we embed the modified ML-RBF into a multi-label Error-Correcting Output Codes (ECOC) method in order to further consider the subcellular localization dependency. We name our method ML-rRBF-ECOC. Finally, the performance of ML-rRBF-ECOC is evaluated on three benchmark datasets.

Results:

The results demonstrate that ML-rRBF-ECOC has highly competitive performance to the related multi-label learning method and some state-of-the-art methods for predicting protein subcellular localizations with multiple sites. Considering dependency between subcellular localizations can contribute to the improvement of prediction performance.

Conclusion:

This also indicates that correlations between different subcellular localizations really exist. Our method at least plays a complementary role to existing methods for predicting protein subcellular localizations with multiple sites.

Functional characterization of an orexin neuropeptide in amphioxus reveals an ancient origin of orexin/orexin receptor system in chordate

Amphioxus belongs to the subphylum cephalochordata, an extant representative of the most basal chordates, whose regulation of endocrine system remains ambiguous. Here we clearly demonstrated the existence of a functional orexin neuropeptide in amphioxus, which is able to interact with orexin receptor, activate both PKC and PKA pathways, decrease leptin expression, and stimulate lipogenesis. We also showed the transcription level of amphioxus orexin was affected by fasting or temperature, indicating a role of this gene in the regulation of energy balance. In addition, the expression of the amphioxus orexin was detected at cerebral vesicle, which has been proposed to be a homolog of the vertebrate brain. These data collectively suggest that a functional orexin neuropeptide has already emerged in amphioxus, which provide insights into the evolutionary origin of orexin in chordate and the functional homology between the cerebral vesicle and vertebrate brain.

Prediction of Protein Subcellular Localization Based on Fusion of Multi-view Features

The prediction of protein subcellular localization is critical for inferring protein functions, gene regulations and protein-protein interactions. With the advances of high-throughput sequencing technologies and proteomic methods, the protein sequences of numerous yeasts have become publicly available, which enables us to computationally predict yeast protein subcellular localization. However, widely-used protein sequence representation techniques, such as amino acid composition and the Chou's pseudo amino acid composition (PseAAC), are difficult in extracting adequate information about the interactions between residues and position distribution of each residue. Therefore, it is still urgent to develop novel sequence representations. In this study, we have presented two novel protein sequence representation techniques including Generalized Chaos Game Representation (GCGR) based on the frequency and distributions of the residues in the protein primary sequence, and novel statistics and information theory (NSI) reflecting local position information of the sequence. In the GCGR + NSI representation, a protein primary sequence is simply represented by a 5-dimensional feature vector, while other popular methods like PseAAC and dipeptide adopt features of more than hundreds of dimensions. In practice, the feature representation is highly efficient in predicting protein subcellular localization. Even without using machine learning-based classifiers, a simple model based on the feature vector can achieve prediction accuracies of 0.8825 and 0.7736 respectively for the CL317 and ZW225 datasets. To further evaluate the effectiveness of the proposed encoding schemes, we introduce a multi-view features-based method to combine the two above-mentioned features with other well-known features including PseAAC and dipeptide composition, and use support vector machine as the classifier to predict protein subcellular localization. This novel model achieves prediction accuracies of 0.927 and 0.871 respectively for the CL317 and ZW225 datasets, better than other existing methods in the jackknife tests. The results suggest that the GCGR and NSI features are useful complements to popular protein sequence representations in predicting yeast protein subcellular localization. Finally, we validate a few newly predicted protein subcellular localizations by evidences from some published articles in authority journals and books.

Efficient computational model for classification of protein localization images using Extended Threshold Adjacency Statistics and Support Vector Machines

BACKGROUND AND OBJECTIVE

Discriminative and informative feature extraction is the core requirement for accurate and efficient classification of protein subcellular localization images so that drug development could be more effective. The objective of this paper is to propose a novel modification in the Threshold Adjacency Statistics technique and enhance its discriminative power.

METHODS

In this work, we utilized Threshold Adjacency Statistics from a novel perspective to enhance its discrimination power and efficiency. In this connection, we utilized seven threshold ranges to produce seven distinct feature spaces, which are then used to train seven SVMs. The final prediction is obtained through the majority voting scheme. The proposed ETAS-SubLoc system is tested on two benchmark datasets using 5-fold cross-validation technique.

RESULTS

We observed that our proposed novel utilization of TAS technique has improved the discriminative power of the classifier. The ETAS-SubLoc system has achieved 99.2% accuracy, 99.3% sensitivity and 99.1% specificity for Endogenous dataset outperforming the classical Threshold Adjacency Statistics technique. Similarly, 91.8% accuracy, 96.3% sensitivity and 91.6% specificity values are achieved for Transfected dataset.

CONCLUSIONS

Simulation results validated the effectiveness of ETAS-SubLoc that provides superior prediction performance compared to the existing technique. The proposed methodology aims at providing support to pharmaceutical industry as well as research community towards better drug designing and innovation in the fields of bioinformatics and computational biology. The implementation code for replicating the experiments presented in this paper is available at: https://drive.google.com/file/d/0B7IyGPObWbSqRTRMcXI2bG5CZWs/view?usp=sharing.

A Pipeline for Classifying Deleterious Coding Mutations in Agricultural Plants

The impact of deleterious variation on both plant fitness and crop productivity is not completely understood and is a hot topic of debates. The deleterious mutations in plants have been solely predicted using sequence conservation methods rather than function-based classifiers due to lack of well-annotated mutational datasets in these organisms. Here, we developed a machine learning classifier based on a dataset of deleterious and neutral mutations in Arabidopsis thaliana by extracting 18 informative features that discriminate deleterious mutations from neutral, including 9 novel features not used in previous studies. We examined linear SVM, Gaussian SVM, and Random Forest classifiers, with the latter performing best. Random Forest classifiers exhibited a markedly higher accuracy than the popular PolyPhen-2 tool in the Arabidopsis dataset. Additionally, we tested whether the Random Forest, trained on the Arabidopsis dataset, accurately predicts deleterious mutations in Orýza sativa and Pisum sativum and observed satisfactory levels of performance accuracy (87% and 93%, respectively) higher than obtained by the PolyPhen-2. Application of Transfer learning in classifiers did not improve their performance. To additionally test the performance of the Random Forest classifier across different angiosperm species, we applied it to annotate deleterious mutations in Cicer arietinum and validated them using population frequency data. Overall, we devised a classifier with the potential to improve the annotation of putative functional mutations in QTL and GWAS hit regions, as well as for the evolutionary analysis of proliferation of deleterious mutations during plant domestication; thus optimizing breeding improvement and development of new cultivars.

Learning from Heterogeneous Data Sources: An Application in Spatial Proteomics

Sub-cellular localisation of proteins is an essential post-translational regulatory mechanism that can be assayed using high-throughput mass spectrometry (MS). These MS-based spatial proteomics experiments enable us to pinpoint the sub-cellular distribution of thousands of proteins in a specific system under controlled conditions. Recent advances in high-throughput MS methods have yielded a plethora of experimental spatial proteomics data for the cell biology community. Yet, there are many third-party data sources, such as immunofluorescence microscopy or protein annotations and sequences, which represent a rich and vast source of complementary information. We present a unique transfer learning classification framework that utilises a nearest-neighbour or support vector machine system, to integrate heterogeneous data sources to considerably improve on the quantity and quality of sub-cellular protein assignment. We demonstrate the utility of our algorithms through evaluation of five experimental datasets, from four different species in conjunction with four different auxiliary data sources to classify proteins to tens of sub-cellular compartments with high generalisation accuracy. We further apply the method to an experiment on pluripotent mouse embryonic stem cells to classify a set of previously unknown proteins, and validate our findings against a recent high resolution map of the mouse stem cell proteome. The methodology is distributed as part of the open-source Bioconductor pRoloc suite for spatial proteomics data analysis.

Sub-cellular localisation of proteins is critical to their function in all cellular processes; proteins localising to their intended micro-environment, e.g organelles, vesicles or macro-molecular complexes, will meet the interaction partners and biochemical conditions suitable to pursue their molecular function. Therefore, sound data and methods to reliably and systematically study protein localisation, and hence their mis-localisation and the disruption of protein trafficking, that are relied upon by the cell biology community, are essential. Here we present a method to infer protein localisation relying on the optimal integration of experimental mass spectrometry-based data and auxiliary sources, such as GO annotation, outputs from third-party software, protein-protein interactions or immunocytochemistry data. We found that the application of transfer learning algorithms across these diverse data sources considerably improves on the quantity and reliability of sub-cellular protein assignment, compared to single data classifiers previously applied to infer sub-cellular localisation using experimental data only. We show how our method does not compromise biologically relevant experimental-specific signal after integration with heterogeneous freely available third-party resources. The integration of different data sources is an important challenge in the data intensive world of biology and we anticipate the transfer learning methods presented here will prove useful to many areas of biology, to unify data obtained from different but complimentary sources.

Multi-label multi-instance transfer learning for simultaneous reconstruction and cross-talk modeling of multiple human signaling pathways

BACKGROUND

Signaling pathways play important roles in the life processes of cell growth, cell apoptosis and organism development. At present the signal transduction networks are far from complete. As an effective complement to experimental methods, computational modeling is suited to rapidly reconstruct the signaling pathways at low cost. To our knowledge, the existing computational methods seldom simultaneously exploit more than three signaling pathways into one predictive model for the discovery of novel signaling components and the cross-talk modeling between signaling pathways.

RESULTS

In this work, we propose a multi-label multi-instance transfer learning method to simultaneously reconstruct 27 human signaling pathways and model their cross-talks. Computational results show that the proposed method demonstrates satisfactory multi-label learning performance and rational proteome-wide predictions. Some predicted signaling components or pathway targeted proteins have been validated by recent literature. The predicted signaling components are further linked to pathways using the experimentally derived PPIs (protein-protein interactions) to reconstruct the human signaling pathways. Thus the map of the cross-talks via common signaling components and common signaling PPIs is conveniently inferred to provide valuable insights into the regulatory and cooperative relationships between signaling pathways. Lastly, gene ontology enrichment analysis is conducted to gain statistical knowledge about the reconstructed human signaling pathways.

CONCLUSIONS

Multi-label learning framework has been demonstrated effective in this work to model the phenomena that a signaling protein belongs to more than one signaling pathway. As results, novel signaling components and pathways targeted proteins are predicted to simultaneously reconstruct multiple human signaling pathways and the static map of their cross-talks for further biomedical research.

AdaBoost based multi-instance transfer learning for predicting proteome-wide interactions between Salmonella and human proteins

Pathogen-host protein-protein interaction (PPI) plays an important role in revealing the underlying pathogenesis of viruses and bacteria. The need of rapidly mapping proteome-wide pathogen-host interactome opens avenues for and imposes burdens on computational modeling. For Salmonella typhimurium, only 62 interactions with human proteins are reported to date, and the computational modeling based on such a small training data is prone to yield model overfitting. In this work, we propose a multi-instance transfer learning method to reconstruct the proteome-wide Salmonella-human PPI networks, wherein the training data is augmented by homolog knowledge transfer in the form of independent homolog instances. We use AdaBoost instance reweighting to counteract the noise from homolog instances, and deliberately design three experimental settings to validate the assumption that the homolog instances are effective to address the problems of data scarcity and data unavailability. The experimental results show that the proposed method outperforms the existing models and some predictions are validated by the findings from recent literature. Lastly, we conduct gene ontology based clustering analysis of the predicted networks to provide insights into the pathogenesis of Salmonella.

CELLO2GO: a web server for protein subCELlular LOcalization prediction with functional gene ontology annotation

CELLO2GO (http://cello.life.nctu.edu.tw/cello2go/) is a publicly available, web-based system for screening various properties of a targeted protein and its subcellular localization. Herein, we describe how this platform is used to obtain a brief or detailed gene ontology (GO)-type categories, including subcellular localization(s), for the queried proteins by combining the CELLO localization-predicting and BLAST homology-searching approaches. Given a query protein sequence, CELLO2GO uses BLAST to search for homologous sequences that are GO annotated in an in-house database derived from the UniProt KnowledgeBase database. At the same time, CELLO attempts predict at least one subcellular localization on the basis of the species in which the protein is found. When homologs for the query sequence have been identified, the number of terms found for each of their GO categories, i.e., cellular compartment, molecular function, and biological process, are summed and presented as pie charts representing possible functional annotations for the queried protein. Although the experimental subcellular localization of a protein may not be known, and thus not annotated, CELLO can confidentially suggest a subcellular localization. CELLO2GO should be a useful tool for research involving complex subcellular systems because it combines CELLO and BLAST into one platform and its output is easily manipulated such that the user-specific questions may be readily addressed.

HybridGO-Loc: mining hybrid features on gene ontology for predicting subcellular localization of multi-location proteins

Protein subcellular localization prediction, as an essential step to elucidate the functions in vivo of proteins and identify drugs targets, has been extensively studied in previous decades. Instead of only determining subcellular localization of single-label proteins, recent studies have focused on predicting both single- and multi-location proteins. Computational methods based on Gene Ontology (GO) have been demonstrated to be superior to methods based on other features. However, existing GO-based methods focus on the occurrences of GO terms and disregard their relationships. This paper proposes a multi-label subcellular-localization predictor, namely HybridGO-Loc, that leverages not only the GO term occurrences but also the inter-term relationships. This is achieved by hybridizing the GO frequencies of occurrences and the semantic similarity between GO terms. Given a protein, a set of GO terms are retrieved by searching against the gene ontology database, using the accession numbers of homologous proteins obtained via BLAST search as the keys. The frequency of GO occurrences and semantic similarity (SS) between GO terms are used to formulate frequency vectors and semantic similarity vectors, respectively, which are subsequently hybridized to construct fusion vectors. An adaptive-decision based multi-label support vector machine (SVM) classifier is proposed to classify the fusion vectors. Experimental results based on recent benchmark datasets and a new dataset containing novel proteins show that the proposed hybrid-feature predictor significantly outperforms predictors based on individual GO features as well as other state-of-the-art predictors. For readers' convenience, the HybridGO-Loc server, which is for predicting virus or plant proteins, is available online at http://bioinfo.eie.polyu.edu.hk/HybridGoServer/.

Robust feature generation for protein subchloroplast location prediction with a weighted GO transfer model

Chloroplasts are crucial organelles of green plants and eukaryotic algae since they conduct photosynthesis. Predicting the subchloroplast location of a protein can provide important insights for understanding its biological functions. The performance of subchloroplast location prediction algorithms often depends on deriving predictive and succinct features from genomic and proteomic data. In this work, a novel weighted Gene Ontology (GO) transfer model is proposed to generate discriminating features from sequence data and GO Categories. This model contains two components. First, we transfer the GO terms of the homologous protein, and then assign the bit-score as weights to GO features. Second, we employ term-selection methods to determine weights for GO terms. This model is capable of improving prediction accuracy due to the tolerance of the noise derived from homolog knowledge transfer. The proposed weighted GO transfer method based on bit-score and a logarithmic transformation of CHI-square (WS-LCHI) performs better than the baseline models, and also outperforms the four off-the-shelf subchloroplast prediction methods.

Probability weighted ensemble transfer learning for predicting interactions between HIV-1 and human proteins

Reconstruction of host-pathogen protein interaction networks is of great significance to reveal the underlying microbic pathogenesis. However, the current experimentally-derived networks are generally small and should be augmented by computational methods for less-biased biological inference. From the point of view of computational modelling, data scarcity, data unavailability and negative data sampling are the three major problems for host-pathogen protein interaction networks reconstruction. In this work, we are motivated to address the three concerns and propose a probability weighted ensemble transfer learning model for HIV-human protein interaction prediction (PWEN-TLM), where support vector machine (SVM) is adopted as the individual classifier of the ensemble model. In the model, data scarcity and data unavailability are tackled by homolog knowledge transfer. The importance of homolog knowledge is measured by the ROC-AUC metric of the individual classifiers, whose outputs are probability weighted to yield the final decision. In addition, we further validate the assumption that only the homolog knowledge is sufficient to train a satisfactory model for host-pathogen protein interaction prediction. Thus the model is more robust against data unavailability with less demanding data constraint. As regards with negative data construction, experiments show that exclusiveness of subcellular co-localized proteins is unbiased and more reliable than random sampling. Last, we conduct analysis of overlapped predictions between our model and the existing models, and apply the model to novel host-pathogen PPIs recognition for further biological research.

Profiling of the mammalian mitotic spindle proteome reveals an ER protein, OSTD-1, as being necessary for cell division and ER morphology

Cell division is important for many cellular processes including cell growth, reproduction, wound healing and stem cell renewal. Failures in cell division can often lead to tumors and birth defects. To identify factors necessary for this process, we implemented a comparative profiling strategy of the published mitotic spindle proteome from our laboratory. Of the candidate mammalian proteins, we determined that 77% had orthologs in Caenorhabditis elegans and 18% were associated with human disease. Of the C. elegans candidates (n=146), we determined that 34 genes functioned in embryonic development and 56% of these were predicted to be membrane trafficking proteins. A secondary, visual screen to detect distinct defects in cell division revealed 21 genes that were necessary for cytokinesis. One of these candidates, OSTD-1, an ER resident protein, was further characterized due to the aberrant cleavage furrow placement and failures in division. We determined that OSTD-1 plays a role in maintaining the dynamic morphology of the ER during the cell cycle. In addition, 65% of all ostd-1 RNAi-treated embryos failed to correctly position cleavage furrows, suggesting that proper ER morphology plays a necessary function during animal cell division.

SVM ensemble based transfer learning for large-scale membrane proteins discrimination

Membrane proteins play important roles in molecular trans-membrane transport, ligand-receptor recognition, cell-cell interaction, enzyme catalysis, host immune defense response and infectious disease pathways. Up to present, discriminating membrane proteins remains a challenging problem from the viewpoints of biological experimental determination and computational modeling. This work presents SVM ensemble based transfer learning model for membrane proteins discrimination (SVM-TLM). To reduce the data constraints on computational modeling, this method investigates the effectiveness of transferring the homolog knowledge to the target membrane proteins under the framework of probability weighted ensemble learning. As compared to multiple kernel learning based transfer learning model, the method takes the advantages of sparseness based SVM optimization on large data, thus more computationally efficient for large protein data analysis. The experiments on large membrane protein benchmark dataset show that SVM-TLM achieves significantly better cross validation performance than the baseline model.

Multilabel learning via random label selection for protein subcellular multilocations prediction

Prediction of protein subcellular localization is an important but challenging problem, particularly when proteins may simultaneously exist at, or move between, two or more different subcellular location sites. Most of the existing protein subcellular localization methods are only used to deal with the single-location proteins. In the past few years, only a few methods have been proposed to tackle proteins with multiple locations. However, they only adopt a simple strategy, that is, transforming the multilocation proteins to multiple proteins with single location, which does not take correlations among different subcellular locations into account. In this paper, a novel method named random label selection (RALS) (multilabel learning via RALS), which extends the simple binary relevance (BR) method, is proposed to learn from multilocation proteins in an effective and efficient way. RALS does not explicitly find the correlations among labels, but rather implicitly attempts to learn the label correlations from data by augmenting original feature space with randomly selected labels as its additional input features. Through the fivefold cross-validation test on a benchmark data set, we demonstrate our proposed method with consideration of label correlations obviously outperforms the baseline BR method without consideration of label correlations, indicating correlations among different subcellular locations really exist and contribute to improvement of prediction performance. Experimental results on two benchmark data sets also show that our proposed methods achieve significantly higher performance than some other state-of-the-art methods in predicting subcellular multilocations of proteins. The prediction web server is available at >http://levis.tongji.edu.cn:8080/bioinfo/MLPred-Euk/ for the public usage.

An ensemble method for predicting subnuclear localizations from primary protein structures

Background

Predicting protein subnuclear localization is a challenging problem. Some previous works based on non-sequence information including Gene Ontology annotations and kernel fusion have respective limitations. The aim of this work is twofold: one is to propose a novel individual feature extraction method; another is to develop an ensemble method to improve prediction performance using comprehensive information represented in the form of high dimensional feature vector obtained by 11 feature extraction methods.

Methodology/Principal Findings

A novel two-stage multiclass support vector machine is proposed to predict protein subnuclear localizations. It only considers those feature extraction methods based on amino acid classifications and physicochemical properties. In order to speed up our system, an automatic search method for the kernel parameter is used. The prediction performance of our method is evaluated on four datasets: Lei dataset, multi-localization dataset, SNL9 dataset and a new independent dataset. The overall accuracy of prediction for 6 localizations on Lei dataset is 75.2% and that for 9 localizations on SNL9 dataset is 72.1% in the leave-one-out cross validation, 71.7% for the multi-localization dataset and 69.8% for the new independent dataset, respectively. Comparisons with those existing methods show that our method performs better for both single-localization and multi-localization proteins and achieves more balanced sensitivities and specificities on large-size and small-size subcellular localizations. The overall accuracy improvements are 4.0% and 4.7% for single-localization proteins and 6.5% for multi-localization proteins. The reliability and stability of our classification model are further confirmed by permutation analysis.

Conclusions

It can be concluded that our method is effective and valuable for predicting protein subnuclear localizations. A web server has been designed to implement the proposed method. It is freely available at http://bioinformatics.awowshop.com/snlpred_page.php.

GOASVM: a subcellular location predictor by incorporating term-frequency gene ontology into the general form of Chou's pseudo-amino acid composition

Prediction of protein subcellular localization is an important yet challenging problem. Recently, several computational methods based on Gene Ontology (GO) have been proposed to tackle this problem and have demonstrated superiority over methods based on other features. Existing GO-based methods, however, do not fully use the GO information. This paper proposes an efficient GO method called GOASVM that exploits the information from the GO term frequencies and distant homologs to represent a protein in the general form of Chou's pseudo-amino acid composition. The method first selects a subset of relevant GO terms to form a GO vector space. Then for each protein, the method uses the accession number (AC) of the protein or the ACs of its homologs to find the number of occurrences of the selected GO terms in the Gene Ontology annotation (GOA) database as a means to construct GO vectors for support vector machines (SVMs) classification. With the advantages of GO term frequencies and a new strategy to incorporate useful homologous information, GOASVM can achieve a prediction accuracy of 72.2% on a new independent test set comprising novel proteins that were added to Swiss-Prot six years later than the creation date of the training set. GOASVM and Supplementary materials are available online at http://bioinfo.eie.polyu.edu.hk/mGoaSvmServer/GOASVM.html.

mGOASVM: Multi-label protein subcellular localization based on gene ontology and support vector machines

Background

Although many computational methods have been developed to predict protein subcellular localization, most of the methods are limited to the prediction of single-location proteins. Multi-location proteins are either not considered or assumed not existing. However, proteins with multiple locations are particularly interesting because they may have special biological functions, which are essential to both basic research and drug discovery.

Results

This paper proposes an efficient multi-label predictor, namely mGOASVM, for predicting the subcellular localization of multi-location proteins. Given a protein, the accession numbers of its homologs are obtained via BLAST search. Then, the original accession number and the homologous accession numbers of the protein are used as keys to search against the Gene Ontology (GO) annotation database to obtain a set of GO terms. Given a set of training proteins, a set of T relevant GO terms is obtained by finding all of the GO terms in the GO annotation database that are relevant to the training proteins. These relevant GO terms then form the basis of a T-dimensional Euclidean space on which the GO vectors lie. A support vector machine (SVM) classifier with a new decision scheme is proposed to classify the multi-label GO vectors. The mGOASVM predictor has the following advantages: (1) it uses the frequency of occurrences of GO terms for feature representation; (2) it selects the relevant GO subspace which can substantially speed up the prediction without compromising performance; and (3) it adopts an efficient multi-label SVM classifier which significantly outperforms other predictors. Briefly, on two recently published virus and plant datasets, mGOASVM achieves an actual accuracy of 88.9% and 87.4%, respectively, which are significantly higher than those achieved by the state-of-the-art predictors such as iLoc-Virus (74.8%) and iLoc-Plant (68.1%).

Conclusions

mGOASVM can efficiently predict the subcellular locations of multi-label proteins. The mGOASVM predictor is available online at http://bioinfo.eie.polyu.edu.hk/mGoaSvmServer/mGOASVM.html.

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.

Imbalanced multi-modal multi-label learning for subcellular localization prediction of human proteins with both single and multiple sites

It is well known that an important step toward understanding the functions of a protein is to determine its subcellular location. Although numerous prediction algorithms have been developed, most of them typically focused on the proteins with only one location. In recent years, researchers have begun to pay attention to the subcellular localization prediction of the proteins with multiple sites. However, almost all the existing approaches have failed to take into account the correlations among the locations caused by the proteins with multiple sites, which may be the important information for improving the prediction accuracy of the proteins with multiple sites. In this paper, a new algorithm which can effectively exploit the correlations among the locations is proposed by using gaussian process model. Besides, the algorithm also can realize optimal linear combination of various feature extraction technologies and could be robust to the imbalanced data set. Experimental results on a human protein data set show that the proposed algorithm is valid and can achieve better performance than the existing approaches.

WegoLoc: accurate prediction of protein subcellular localization using weighted Gene Ontology terms

SUMMARY

We present an accurate and fast web server, WegoLoc for predicting subcellular localization of proteins based on sequence similarity and weighted Gene Ontology (GO) information. A term weighting method in the text categorization process is applied to GO terms for a support vector machine classifier. As a result, WegoLoc surpasses the state-of-the-art methods for previously used test datasets. WegoLoc supports three eukaryotic kingdoms (animals, fungi and plants) and provides human-specific analysis, and covers several sets of cellular locations. In addition, WegoLoc provides (i) multiple possible localizations of input protein(s) as well as their corresponding probability scores, (ii) weights of GO terms representing the contribution of each GO term in the prediction, and (iii) a BLAST E-value for the best hit with GO terms. If the similarity score does not meet a given threshold, an amino acid composition-based prediction is applied as a backup method.

AVAILABILITY

WegoLoc and User's guide are freely available at the website http://www.btool.org/WegoLoc

CONTACT

smchiks@ks.ac.kr; dougnam@unist.ac.kr

SUPPLEMENTARY INFORMATION

Supplementary data is available at http://www.btool.org/WegoLoc.

An ensemble classifier for eukaryotic protein subcellular location prediction using gene ontology categories and amino acid hydrophobicity

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.

CE-PLoc: an ensemble classifier for predicting protein subcellular locations by fusing different modes of pseudo amino acid composition

Precise information about protein locations in a cell facilitates in the understanding of the function of a protein and its interaction in the cellular environment. This information further helps in the study of the specific metabolic pathways and other biological processes. We propose an ensemble approach called "CE-PLoc" for predicting subcellular locations based on fusion of individual classifiers. The proposed approach utilizes features obtained from both dipeptide composition (DC) and amphiphilic pseudo amino acid composition (PseAAC) based feature extraction strategies. Different feature spaces are obtained by varying the dimensionality using PseAAC for a selected base learner. The performance of the individual learning mechanisms such as support vector machine, nearest neighbor, probabilistic neural network, covariant discriminant, which are trained using PseAAC based features is first analyzed. Classifiers are developed using same learning mechanism but trained on PseAAC based feature spaces of varying dimensions. These classifiers are combined through voting strategy and an improvement in prediction performance is achieved. Prediction performance is further enhanced by developing CE-PLoc through the combination of different learning mechanisms trained on both DC based feature space and PseAAC based feature spaces of varying dimensions. The predictive performance of proposed CE-PLoc is evaluated for two benchmark datasets of protein subcellular locations using accuracy, MCC, and Q-statistics. Using the jackknife test, prediction accuracies of 81.47 and 83.99% are obtained for 12 and 14 subcellular locations datasets, respectively. In case of independent dataset test, prediction accuracies are 87.04 and 87.33% for 12 and 14 class datasets, respectively.