Search

When it comes to biological experiments, especially the experiments related to protein and drug binding, thorough experiments are not feasible because they will cost a lot of manpower and material resources [1]. Therefore, it has become a popular method to select a series of experiments to be carried out and effectively learns the model to predict the results of unfinished experiments, that is, active learning. Based on the existing experimental data, this paper discusses the feasibility of machine learning in biological experiments. Ordinary supervised learning and active learning are used to build prediction models, respectively. The difference between them is that active learning purposefully selects the "most useful" data for multiple rounds of learning, which is more in line with the needs of actual experiments. The result is that the accuracy of active learning is slightly higher than that of ordinary supervised learning when almost one fifth of the data of ordinary supervised learning is used.

Analysis of protein subcellular localization is a critical part of proteomics. In recent years, as both the number and quality of microscopic images are increasing rapidly, many automated methods, especially convolutional neural networks (CNN), have been developed to predict protein subcellular location(s) based on bioimages, but their performance always suffers from some inherent properties of the problem. First, many microscopic images have non-informative or noisy sections, like unstained stroma and unspecific background, which affect the extraction of protein expression information. Second, the patterns of protein subcellular localization are very complex, as a lot of proteins locate in more than one compartment. In this study, we propose a new label-correlation enhanced deep neural network, laceDNN, to classify the subcellular locations of multi-label proteins from immunohistochemistry images. The model uses small representative patches as input to alleviate the image noise issue, and its backbone is a hybrid architecture of CNN and recurrent neural network, where the former network extracts representative image features and the latter learns the organelle dependency relationships. Our experimental results indicate that the proposed model can improve the performance of multi-label protein subcellular classification.

The spatial distribution of proteome at subcellular levels provides clues for protein functions, thus is important to human biology and medicine. Imaging-based methods are one of the most important approaches for predicting protein subcellular location. Although deep neural networks have shown impressive performance in a number of imaging tasks, its application to protein subcellular localization has not been sufficiently explored. In this study, we developed a deep imaging-based approach to localize the proteins at subcellular levels. Based on deep image features extracted from convolutional neural networks (CNNs), both single-label and multi-label locations can be accurately predicted. Particularly, the multi-label prediction is quite a challenging task. Here we developed a criterion learning strategy to exploit the label–attribute relevancy and label–label relevancy. A criterion that was used to determine the final label set was automatically obtained during the learning procedure. We concluded an optimal CNN architecture that could give the best results. Besides, experiments show that compared with the hand-crafted features, the deep features present more accurate prediction with less features. The implementation for the proposed method is available at https://github.com/RanSuLab/ProteinSubcellularLocation.

Motivation For in-depth understanding the functions of proteins in a cell, the knowledge of their subcellular localization is indispensable. The current study is focused on human protein subcellular location prediction based on the sequence information alone. Although considerable efforts have been made in this regard, the problem is far from being solved yet. Most existing methods can be used to deal with single-location proteins only. Actually, proteins with multi-locations may have some special biological functions that are particularly important for both basic research and drug design. Results Using the multi-label theory, we present a new predictor called ‘pLoc-mHum’ by extracting the crucial GO (Gene Ontology) information into the general PseAAC (Pseudo Amino Acid Composition). Rigorous cross-validations on a same stringent benchmark dataset have indicated that the proposed pLoc-mHum predictor is remarkably superior to iLoc-Hum, the state-of-the-art method in predicting the human protein subcellular localization. Availability and implementation To maximize the convenience of most experimental scientists, a user-friendly web-server for the new predictor has been established at http://www.jci-bioinfo.cn/pLoc-mHum/, by which users can easily get their desired results without the need to go through the complicated mathematics involved. Supplementary information Supplementary data are available at Bioinformatics online.

Functions of proteins are in general related to their subcellular locations. To identify the functions of a protein, we first need know where this protein is located. Interacting proteins tend to locate in the same subcellular location. Thus, it is imperative to take the protein-protein interactions into account for computational identification of protein subcellular locations.In this study, we present a deep learning-based method, node2loc, to predict protein subcellular location. node2loc first learns distributed representations of proteins in a protein-protein network using node2vec, which acquires representations from unlabeled data for downstream tasks. Then the learned representations are further fed into a recurrent neural network (RNN) to predict subcellular locations. Considering the severe class imbalance of different subcellular locations, Synthetic Minority Over-sampling Technique (SMOTE) is applied to artificially boost subcellular locations with few proteins.We construct a benchmark dataset with 16 subcellular locations and evaluate node2loc on this dataset. node2loc yields a Matthews correlation coefficient (MCC) value of 0.812, which outperforms other baseline methods. The results demonstrate that the learned presentations from a protein-protein network have strong discriminate ability for classifying protein subcellular locations and the RNN is a more powerful classifier than traditional machine learning models. node2loc is freely available at https://github.com/xypan1232/node2loc.

BackgroundProtein subcellular localization plays a crucial role in understanding cell function. Proteins need to be in the right place at the right time, and combine with the corresponding molecules to fulfill their functions. Furthermore, prediction of protein subcellular location not only should be a guiding role in drug design and development due to potential molecular targets but also be an essential role in genome annotation. Taking the current status of image-based protein subcellular localization as an example, there are three common drawbacks, i.e., obsolete datasets without updating label information, stereotypical feature descriptor on spatial domain or grey level, and single-function prediction algorithm’s limited capacity of handling single-label database.ResultsIn this paper, a novel human protein subcellular localization prediction model MIC_Locator is proposed. Firstly, the latest datasets are collected and collated as our benchmark dataset instead of obsolete data while training prediction model. Secondly, Fourier transformation, Riesz transformation, Log-Gabor filter and intensity coding strategy are employed to obtain frequency feature based on three components of monogenic signal with different frequency scales. Thirdly, a chained prediction model is proposed to handle multi-label instead of single-label datasets. The experiment results showed that the MIC_Locator can achieve 60.56% subset accuracy and outperform the existing majority of prediction models, and the frequency feature and intensity coding strategy can be conducive to improving the classification accuracy.ConclusionsOur results demonstrate that the frequency feature is more beneficial for improving the performance of model compared to features extracted from spatial domain, and the MIC_Locator proposed in this paper can speed up validation of protein annotation, knowledge of protein function and proteomics research.

Protein subcellular location has a huge positive influence on understanding protein function. In the past decades, many image-based automated approaches have been published for predicting protein subcellular location. However, in the reported literatures, there is a common deficiency for diverse prediction models in capturing local information of interest region of image. It motivates us to propose a novel approach by employing local feature descriptor named the Local Tetra Patterns (LTrP). In this paper, local features together with global features were fed to support vector machine to train chain classifiers, which can deal with multi-label datasets by using problem transformation pattern. To verify the validity of our approach, three different experiments were conducted based on the same benchmark dataset. The results show that the performance of the classification with LTrP descriptor not only captured more local information in interest region of images but also contributed to the improvement of prediction precision since the local descriptor is encoded along horizontal and vertical directions by LTrP. By applying the new approach, a more accurate classifier of protein subcellular location can be modeled, which is crucial to screen cancer biomarkers and research pathology mechanisms.

The reproductive system is a specific system of organs working together for the purpose of reproduction. As one of the most significant characteristics of human cell, subcellular localization plays a critical role for understanding specific functions of mammalian proteins. In this study, we developed a novel computational protocol for predicting protein subcellular locations from microscope images of cells in human reproductive tissues. Three major steps are contained in this protocol, i.e., protein object identification, image feature extraction, and classification. We first separated protein and DNA staining in the images with both linear and non-negative matrix factorization separation methods; then we extracted protein multi-view global and local texture features including wavelet Haralick, local binary patterns, local ternary patterns, and the local quinary patterns; finally based on the selected important feature subset, we constructed an ensemble classifier with support vector machines for classifications. Experiments are performed on a benchmark dataset consisting of seven major subcellular classes in human reproductive tissues collected from human protein atlas. Our results show that the local texture pattern features play an important complementary role to global features for enhancing the prediction performance. An overall accuracy of 85% is obtained through current system, and when only confident classifications are considered, the accuracy can reach 99%. It is the first developed image based protein subcellular location predictor specifically for human reproductive tissue. The promising results indicate that the developed protocol can be applied for accurate large-scale protein subcellular localization annotations in human reproductive system.

Motivation Protein subcellular localization prediction has been an important research topic in computational biology over the last decade. Various automatic methods have been proposed to predict locations for large scale protein datasets, where statistical machine learning algorithms are widely used for model construction. A key step in these predictors is encoding the amino acid sequences into feature vectors. Many studies have shown that features extracted from biological domains, such as gene ontology and functional domains, can be very useful for improving the prediction accuracy. However, domain knowledge usually results in redundant features and high-dimensional feature spaces, which may degenerate the performance of machine learning models. Results In this paper, we propose a new amino acid sequence-based human protein subcellular location prediction approach Hum-mPLoc 3.0, which covers 12 human subcellular localizations. The sequences are represented by multi-view complementary features, i.e. context vocabulary annotation-based gene ontology (GO) terms, peptide-based functional domains, and residue-based statistical features. To systematically reflect the structural hierarchy of the domain knowledge bases, we propose a novel feature representation protocol denoted as HCM (Hidden Correlation Modeling), which will create more compact and discriminative feature vectors by modeling the hidden correlations between annotation terms. Experimental results on four benchmark datasets show that HCM improves prediction accuracy by 5–11% and F1 by 8–19% compared with conventional GO-based methods. A large-scale application of Hum-mPLoc 3.0 on the whole human proteome reveals proteins co-localization preferences in the cell. Availability and Implementation www.csbio.sjtu.edu.cn/bioinf/Hum-mPLoc3/ Supplementary information Supplementary data are available at Bioinformatics online.

Predicting protein subcellular location is necessary for understanding cell function. Several machine learning methods have been developed for computational prediction of primary protein sequences because wet experiments are costly and time consuming. However, two problems still exist in state-of-the-art methods. First, several proteins appear in different subcellular structures simultaneously, whereas current methods only predict one protein sequence in one subcellular structure. Second, most software tools are trained with obsolete data and the latest new databases are missed. We proposed a novel multi-label classification algorithm to solve the first problem and integrated several latest databases to improve prediction performance. Experiments proved the effectiveness of the proposed method. The present study would facilitate research on cellular proteomics.

High throughput screening determines the effects of many conditions on a given biological target. Currently, to estimate the effects of those conditions on other targets requires either strong modeling assumptions (e.g. similarities among targets) or separate screens. Ideally, data-driven experimentation could be used to learn accurate models for many conditions and targets without doing all possible experiments. We have previously described an active machine learning algorithm that can iteratively choose small sets of experiments to learn models of multiple effects. We now show that, with no prior knowledge and with liquid handling robotics and automated microscopy under its control, this learner accurately learned the effects of 48 chemical compounds on the subcellular localization of 48 proteins while performing only 29% of all possible experiments. The results represent the first practical demonstration of the utility of active learning-driven biological experimentation in which the set of possible phenotypes is unknown in advance. DOI: http://dx.doi.org/10.7554/eLife.10047.001, eLife digest Biomedical scientists have invested significant effort into making it easy to perform lots of experiments quickly and cheaply. These “high throughput” methods are the workhorses of modern “systems biology” efforts. However, we simply cannot perform an experiment for every possible combination of different cell type, genetic mutation and other conditions. In practice this has led researchers to either exhaustively test a few conditions or targets, or to try to pick the experiments that best allow a particular problem to be explored. But which experiments should we pick? The ones we think we can predict the outcome of accurately, the ones for which we are uncertain what the results will be, or a combination of the two? Humans are not particularly well suited for this task because it requires reasoning about many possible outcomes at the same time. However, computers are much better at handling statistics for many experiments, and machine learning algorithms allow computers to “learn” how to make predictions and decisions based on the data they’ve previously processed. Previous computer simulations showed that a machine learning approach termed “active learning” could do a good job of picking a series of experiments to perform in order to efficiently learn a model that predicts the results of experiments that were not done. Now, Naik et al. have performed cell biology experiments in which experiments were chosen by an active learning algorithm and then performed using liquid handling robots and an automated microscope. The key idea behind the approach is that you learn more from an experiment you can’t predict (or that you predicted incorrectly) than from just confirming your confident predictions. The results of the robot-driven experiments showed that the active learning approach outperforms strategies a human might use, even when the potential outcomes of individual experiments are not known beforehand. The next challenge is to apply these methods to reduce the cost of achieving the goals of large projects, such as The Cancer Genome Atlas. DOI: http://dx.doi.org/10.7554/eLife.10047.002

Motivation: We present a framework and algorithms to intelligently acquire movies of protein subcellular location patterns by learning their models as they are being acquired, and simultaneously determining how many cells to acquire as well as how many frames to acquire per cell. This is motivated by the desire to minimize acquisition time and photobleaching, given the need to build such models for all proteins, in all cell types, under all conditions. Our key innovation is to build models during acquisition rather than as a post-processing step, thus allowing us to intelligently and automatically adapt the acquisition process given the model acquired. Results: We validate our framework on protein subcellular location classification, and show that the combination of model building and intelligent acquisition results in time and storage savings without loss of classification accuracy, or alternatively, higher classification accuracy for the same total acquisition time. Availability and implementation: The data and software used for this study will be made available upon publication at http://murphylab.web.cmu.edu/software and http://www.andrew.cmu.edu/user/jelenak/Software. Contact: jelenak@cmu.edu

Optimally weighted fuzzy k-nearest neighbors (OWFKNN) algorithm has been used to predict proteins' subcellular locations based on their amino acid composition, in this paper. The datasets used consists of two species which are 997 prokaryotic and 2427 eukaryotic protein sequences. The overall prediction accuracy achieved is about 88.5% for prokaryotic sequences and 86.2% for eukaryotic sequences in a jackknife test. Compared to other algorithms developed for the prediction of protein subcellular location, OWFKNN gives very satisfying results. Therefore, OWFKNN can be used as an alternative method to predict protein localization. (C) 2008 Elsevier Ltd. All rights reserved.

Fluorescence microscopy is a powerful tool for analyzing the subcellular distributions of proteins, but that power has not been fully utilized because most analysis of those distributions has been done by visual examination. This limitation can be overcome using automated pattern recognition methods widely used in other fields. This article summarizes work demonstrating that automated systems can recognize the patterns of major organelles in both two- and three-dimensional images of cultured cells, and that these systems can distinguish similar patterns better than visual examination. The basis for these systems are sets of Subcellular Location Features that capture the essence of subcellular patterns without being sensitive to the extensive variation that occurs in the size, shape, and orientation of cells in microscope images. These features can also be used to make sensitive, statistical comparisons of the distribution of a protein between two conditions, such as in the presence and absence of a drug. The possible use of automated pattern analysis methods for improving detection of abnormal cells in cancerous or precancerous tissues is also discussed.

The ongoing biotechnology revolution promises a complete understanding of the mechanisms by which cells and tissues carry out their functions. Central to that goal is the determination of the function of each protein that is present in a given cell type, and determining a protein's location within cells is critical to understanding its function. As large amounts of data become available from genome-wide determination of protein subcellular location, automated approaches to categorizing and comparing location patterns are urgently needed. Since sub-cellular location is most often determined using fluorescence microscopy, we have developed automated systems for interpreting the resulting images. We report here improved numeric features for describing such images that are fairly robust to image intensity binning and spatial resolution. We validate these features by using them to train neural networks that accurately recognize all major subcellular patterns with an accuracy higher than any previously reported. Having validated the features by using them for classification, we also demonstrate using them to create Subcellular Location Trees that group similar proteins and provide a systematic framework for describing subcellular location.

Predicting the subcellular localization of proteins is an important and challenging problem. Traditional experimental approaches are often expensive and time-consuming. Consequently, a growing number of research efforts employ a series of machine learning approaches to predict the subcellular location of proteins. There are two main challenges among the state-of-the-art prediction methods. First, most of the existing techniques are designed to deal with multi-class rather than multi-label classification, which ignores connections between multiple labels. In reality, multiple locations of particular proteins imply that there are vital and unique biological significances that deserve special focus and cannot be ignored. Second, techniques for handling imbalanced data in multi-label classification problems are necessary, but never employed. For solving these two issues, we have developed an ensemble multi-label classifier called HPSLPred, which can be applied for multi-label classification with an imbalanced protein source. For convenience, a user-friendly webserver has been established at http://server.malab.cn/HPSLPred.

Motivation: Knowledge of the subcellular location of a protein is crucial for understanding its functions. The subcellular pattern of a protein is typically represented as the set of cellular components in which it is located, and an important task is to determine this set from microscope images. In this article, we address this classification problem using confocal immunofluorescence images from the Human Protein Atlas (HPA) project. The HPA contains images of cells stained for many proteins; each is also stained for three reference components, but there are many other components that are invisible. Given one such cell, the task is to classify the pattern type of the stained protein. We first randomly select local image regions within the cells, and then extract various carefully designed features from these regions. This region-based approach enables us to explicitly study the relationship between proteins and different cell components, as well as the interactions between these components. To achieve these two goals, we propose two discriminative models that extend logistic regression with structured latent variables. The first model allows the same protein pattern class to be expressed differently according to the underlying components in different regions. The second model further captures the spatial dependencies between the components within the same cell so that we can better infer these components. To learn these models, we propose a fast approximate algorithm for inference, and then use gradient-based methods to maximize the data likelihood. Results: In the experiments, we show that the proposed models help improve the classification accuracies on synthetic data and real cellular images. The best overall accuracy we report in this article for classifying 942 proteins into 13 classes of patterns is about 84.6%, which to our knowledge is the best so far. In addition, the dependencies learned are consistent with prior knowledge of cell organization. Availability: http://murphylab.web.cmu.edu/software/. Contact: Jeff.Schneider@cs.cmu.edu, murphy@cmu.edu

Protein subcellular localization aims at predicting the location of a protein within a cell using computational methods. Knowledge of subcellular localization of proteins indicates protein functions and helps in identifying drug targets. 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. To better reflect the characteristics of multiplex proteins, we formulate prediction of subcellular localization of multiplex proteins as a multilabel learning problem. We present and compare two multilabel learning approaches, which exploit correlations between labels and leverage label-specific features, respectively, to induce a high quality prediction model. Experimental results on six protein data sets under various organisms show that our described methods achieve significantly higher performance than any of the existing methods. Among the different multilabel learning methods, we find that methods exploiting label correlations performs better than those leveraging label-specific features.

Protein sub cellular localization aims at predicting the location of a protein within a cell using computational methods. Knowledge of sub cellular localization of proteins indicates protein functions and helps in identifying drug targets. Prediction of protein sub cellular localization is an important but challenging problem, particularly when proteins may simultaneously exist at, or move between, two or more different sub cellular location sites. Most of the existing protein subcellular localization methods are only used to deal with the single-location proteins. To better reflect the characteristics of multiplex proteins, we formulate prediction of subcellular localization of multiplex proteins as a multi-label learning problem. We present and compare two multi-label learning approaches, which exploit correlations between labels and leverage label-specific features, respectively, to induce a high quality prediction model. Experimental results on six protein data sets under various organisms show that our described methods achieve significantly higher performance than any of the existing methods. Among the different multi-label learning methods, we find that methods exploiting label correlations performs better than those leveraging label-specific features.

In bioinformatics fields, Predicting protein subcellular location is an important task, because protein has to be located in its proper position in a cell to perform its biological functions. Therefore, predicting protein location is an important and challenging task in current molecular and cellular biology. In this paper, a computational method based AdaBoost.M1 algorithm and pseudo amino acids composition (PseAAC) to identify protein subcellular location. AdaBoost.M1, an improved algorithm directly extends the original AdaBoost algorithm to the multi-class case without reducing it to multiple two-class problems, is applied to predict the protein subcellular location. In some previous studies conventional amino acid composition is applied to represent a protein. In order to take into account sequence order effects, in this study we use PseAAC that was proposed by Chou instead of convention amino acids composition to represent a protein. To demonstrate AdaBoost.M1 is a robust and efficient model in predicting location, the same protein dataset that was used cedano et al. in 1997 is adopted by us in this paper. From the result, we can draw a conclusion that the accuracy of this method is outperformed than other methods used by previous researchers and can make the prediction into practice.

In this chapter we present an ensemble classifier that performs multi-class classification by combining several kernel classifiers through Decision Direct Acyclic Graph (DDAG). Each base classifier, called K-TIPCAC, is mainly based on the projection of the given points on the Fisher subspace, estimated on the training data, by means of a novel technique. The proposed multiclass classifier is applied to the task of protein subcellular location prediction, which is one of the most difficult multiclass prediction problems in modern computational biology. Although many methods have been proposed in the literature to solve this problem all the existing approaches are affected by some limitations, so that the problem is still open. Experimental results clearly indicate that the proposed technique, called DDAG K-TIPCAC, performs equally, if not better, than state of the art ensemble methods aimed at multi-class classification of highly unbalanced data.

Motivation: Each protein performs its functions within some specific locations in a cell. This subcellular location is important for understanding protein function and for facilitating its purification. There are now many computational techniques for predicting location based on sequence analysis and database information from homologs. A few recent techniques use text from biological abstracts: our goal is to improve the prediction accuracy of such text-based techniques. We identify three techniques for improving text-based prediction: a rule for ambiguous abstract removal, a mechanism for using synonyms from the Gene Ontology (GO) and a mechanism for using the GO hierarchy to generalize terms. We show that these three techniques can significantly improve the accuracy of protein subcellular location predictors that use text extracted from PubMed abstracts whose references are recorded in Swiss-Prot. Contact: duane@cs.ualberta.ca Supplementary information: Supplementary data are available at Bioinformatics online.

Motivation: Predicting the subcellular location of proteins is an active research area, as a protein’s location within the cell provides meaningful cues about its function. Several previous experiments in utilizing text for protein subcellular location prediction, varied in methods, applicability and performance level. In an earlier work we have used a preliminary text classification system and focused on the integration of text features into a sequence-based classifier to improve location prediction performance. Results: Here the focus shifts to the text-based component itself. We introduce EpiLoc, a comprehensive text-based localization system. We provide an in-depth study of textfeature selection, and study several new ways to associate text with proteins, so that textbased location prediction can be performed for practically any protein. We show that EpiLoc’s performance is comparable to (and may even exceed) that of state-of-the-art sequence-based systems. EpiLoc is available at: http://epiloc.cs.queensu.ca.

Detailed knowledge of the subcellular location of all proteins and how they change under various conditions is essential for systems biology efforts to recreate the behavior of cells and organisms. Systematic study of sub-cellular patterns requires automated methods to determine the location pattern for each protein and how it relates to others. Our group has designed sets of numerical features that characterize the location patterns in high-resolution fluorescence microscope images, has shown that these can be used to distinguish patterns better than visual examination, and has used them to automatically group proteins by their patterns. In the current study, we sought to extend our approaches to images obtained from different cell types, microscopy techniques and resolutions. The results indicate that 1) transformation of subcellular location features can be performed so that similar patterns from different cell types are grouped by automated clustering; and 2) there are several basic location patterns whose recognition is insensitive to image resolution over a wide range. The results suggest strategies to be used for collecting and analyzing images from different cell types and with different resolutions.

Facing the explosion of newly generated protein sequences in the post genomic era, we are challenged to develop an automated method for fast and reliably annotating their subcellular locations. Knowledge of subcellular locations of proteins can provide useful hints for revealing their functions and understanding how they interact with each other in cellular networking. Unfortunately, it is both expensive and time-consuming to determine the localization of an uncharacterized protein in a living cell purely based on experiments. To tackle the challenge, a novel hybridization classifier was developed by fusing many basic individual classifiers through a voting system. The "engine" of these basic classifiers was operated by the OET-KNN (Optimized Evidence-Theoretic K-Nearest Neighbor) rule. As a demonstration, predictions were performed with the fusion classifier for proteins among the following 16 localizations: (1) cell wall, (2) centriole, (3) chloroplast, (4) cyanelle, (5) cytoplasm, (6) cytoskeleton, (7) endoplasmic reticulum, (8) extracell, (9) Golgi apparatus, (10) lysosome, (11) mitochondria, (12) nucleus, (13) peroxisome, (14) plasma membrane, (15) plastid, and (16) vacuole. To get rid of redundancy and homology bias, none of the proteins investigated here had/=25% sequence identity to any other in a same subcellular location. The overall success rates thus obtained via the jack-knife cross-validation test and independent dataset test were 81.6% and 83.7%, respectively, which were 46 approximately 63% higher than those performed by the other existing methods on the same benchmark datasets. Also, it is clearly elucidated that the overwhelmingly high success rates obtained by the fusion classifier is by no means a trivial utilization of the GO annotations as prone to be misinterpreted because there is a huge number of proteins with given accession numbers and the corresponding GO numbers, but their subcellular locations are still unknown, and that the percentage of proteins with GO annotations indicating their subcellular components is even less than the percentage of proteins with known subcellular location annotation in the Swiss-Prot database. It is anticipated that the powerful fusion classifier may also become a very useful high throughput tool in characterizing other attributes of proteins according to their sequences, such as enzyme class, membrane protein type, and nuclear receptor subfamily, among many others. A web server, called "Euk-OET-PLoc", has been designed at http://202.120.37.186/bioinf/euk-oet for public to predict subcellular locations of eukaryotic proteins by the fusion OET-KNN classifier.

A common task in cell and molecular biology is to evaluate the difference or similarity among location patterns under different circumstances. Our previous work has described an automated method to objectively compare any pair of subcellular location patterns using well-characterized numerical features. This paper describes an improved version of the previous comparison method by applying nonparametric testing and multiple testing methods. We show that the new approach has better performance for detecting differences, especially on small amounts of data. The new approach was also used to compare location patterns in a dataset containing 3D images. To demonstrate the application of the method, we collected images of drug-treated cells and then use the method to evaluate the effects of drugs on location patterns.

Background Knowledge of the subcellular location of a protein is critical to understanding how that protein works in a cell. This location is frequently determined by the interpretation of fluorescence microscope images. In recent years, automated systems have been developed for consistent and objective interpretation of such images so that the protein pattern in a single cell can be assigned to a known location category. While these systems perform with nearly perfect accuracy for single cell images of all major subcellular structures, their ability to distinguish subpatterns of an organelle (such as two Golgi proteins) is not perfect. Our goal in the work described here was to improve the ability of an automated system to decide which of two similar patterns is present in a field of cells by considering more than one cell at a time. Since cells displaying the same location pattern are often clustered together, considering multiple cells may be expected to improve discrimination between similar patterns. Results We describe how to take advantage of information on experimental conditions to construct a graphical representation for multiple cells in a field. Assuming that a field is composed of a small number of classes, the classification accuracy can be improved by allowing the computed probability of each pattern for each cell to be influenced by the probabilities of its neighboring cells in the model. We describe a novel way to allow this influence to occur, in which we adjust the prior probabilities of each class to reflect the patterns that are present. When this graphical model approach is used on synthetic multi-cell images in which the true class of each cell is known, we observe that the ability to distinguish similar classes is improved without suffering any degradation in ability to distinguish dissimilar classes. The computational complexity of the method is sufficiently low that improved assignments of classes can be obtained for fields of twelve cells in under 0.04 second on a 1600 megahertz processor. Conclusion We demonstrate that graphical models can be used to improve the accuracy of classification of subcellular patterns in multi-cell fluorescence microscope images. We also describe a novel algorithm for inferring classes from a graphical model. The performance and speed suggest that the method will be particularly valuable for analysis of images from high-throughput microscopy. We also anticipate that it will be useful for analyzing the mixtures of cell types typically present in images of tissues. Lastly, we anticipate that the method can be generalized to other problems.

Proteomics is a major current focus of biomedical research, and location proteomics is the important branch of proteomics that systematically studies the subcellular distributions for all proteins expressed in a given cell type. Fluorescence microscopy of labeled proteins is currently the main methodology to obtain location information. Traditionally, microscope images are analyzed by visual inspection, which suffers from inefficiency and inconsistency. Automated and objective interpretation approaches are therefore needed for location proteomics. In this article, we briefly review recent advances in automated imaging interpretation tools, including supervised classification (which assigns location pattern labels to previously unseen images), unsupervised clustering (which groups proteins based on the similarity among their subcellular distributions), and additional statistical tools that can aid cell and molecular biologists who use microscopy in their work.

Confocal microscopy has been widely used for biological studies because of its unique ability to acquire images free of most out-of-focus light. This property is critical for the acquisition of high-quality three-dimensional (3D) images of cells. Automated image analysis and data mining methods can be used to extract rich information contained in such images. In this chapter, we focus on describing effective methods for automatically interpreting 3D images of protein subcellular location patterns.

Knowledge of a protein's subcellular location is essential to a complete understanding of its functions. Automated interpretation methods for protein location patterns are needed for proteomics projects, and we have previously described systems for classifying the major subcellular patterns in cultured mammalian cells. We describe here the calculation of improved 3D Haralick texture features, which yielded a near-perfect classification accuracy when combined with 3D morphological and edge features. In particular, a set of 7 features achieved 98% overall accuracy for classifying 10 major subcellular location patterns in HeLa cells.

In this paper,we constructed a data set of rice proteins with known locations from SWISS-PROT,using the Support Vector Machine to predicte the type of a given rice protein by incorporating sequence information with physics chemistry property of amino acid. Results are assessed through 5-fold cross-validation tests.

The ongoing biotechnology revolution promises a complete understanding of the mechanisms by which cells and tissues carry out their functions. Central to that goal is the determination of the function of each protein that is present in a given cell type, and determining a protein's location within cells is critical to understanding its function. As large amounts of data become available from genome-wide determination of protein subcellular location, automated approaches to categorizing and comparing location patterns are urgently needed. Since subcellular location is most often determined using fluorescence microscopy, we have developed automated systems for interpreting the resulting images. We report here improved numeric features for describing such images that are fairly robust to image intensity binning and spatial resolution. We validate these features by using them to train neural networks that accurately recognize all major subcellular patterns with an accuracy higher than previously reported. Having validated the features by using them for classification, we also demonstrate using them to create Subcellular Location Trees that group similar proteins and provide a systematic framework for describing subcellular location.

Human reproductive system is a unique organ system owing to which humans are capable of reproducing and bearing live offsprings. From a microscopic point of view, this system process requires protein appearing on the right subcellular location at the right time. In this paper, we developed a novel protocol for protein subcellular localisation prediction from human reproductive normal tissues. According to experimental results, three conclusions can be summarised. First, the completed local binary pattern is more discriminative for describing immunohistochemistry images. Second, the proposed ensemble classifier based on support vector machine learning models has a significant improvement. Third, through three different statistical voting approaches, two proteins for male and two proteins for female were identified as the biomarkers in reproductive tissue. These promising results indicate that the developed protocol can be applied not only for accurate large-scale image-based protein subcellular localisation annotations but also for biomarker identification of human reproductive tissue.

There is extensive interest in automating the collection, organization and analysis of biological data. Data in the form of images present special challenges for such efforts. Since fluorescence microscope images are a primary source of information about the location of proteins within cells, we have set as a long-term goal the building of a knowledge base system that can interpret such images in online journals. To this end, we first developed a robot that searches online journals and finds fluorescence microscope images of individual cells. We then characterized the applicability of pattern classification methods we have previously used on images obtained under controlled conditions to images from different sources and to images subjected to manipulations commonly performed during publication. The results indicate the feasibility of developing search engines to find fluorescence microscope images depicting particular subcellular patterns.

Summary: Based on SOAP(Simple Object Access Protocol) technology, the SubLoc server/client suite offers a user-friendly interface for searching and predicting protein subcellular location. Availability:SubLoc SOAP server is available through HTTP protocol at http://www.bioinfo.tsinghua.edu.cn/~tigerchen/cgi-bin/subloc_soap.pl. The client software can be downloaded at http://www.bioinfo.tsinghua.edu.cn/~guotao/soapclient.html. More information is given at http://www.bioinfo.tsinghua.edu.cn/~guotao/soapserver.html. Contact: sunzhr@mail.tsinghua.edu.cn Supplementary information: Supplementary data are available at Bioinformatics online.

Predicting protein subcellular locations may help us understand protein functions and analyse protein interactions with other molecules. Many machine learning and computational techniques have been used to predict protein subcellular locations. In this paper, we propose a new hybrid classification system called SVM-ANFIS based on Support Vector Machines and Adaptive Neuro Fuzzy Inference System for protein subcellular location prediction. The experimental results show that the new system can not only achieve high total accuracies but also improve local accuracies in protein subcellular location prediction.

Background Fluorescence microscopy is widely used to determine the subcellular location of proteins. Efforts to determine location on a proteome-wide basis create a need for automated methods to analyze the resulting images. Over the past ten years, the feasibility of using machine learning methods to recognize all major subcellular location patterns has been convincingly demonstrated, using diverse feature sets and classifiers. On a well-studied data set of 2D HeLa single-cell images, the best performance to date, 91.5%, was obtained by including a set of multiresolution features. This demonstrates the value of multiresolution approaches to this important problem. Results We report here a novel approach for the classification of subcellular location patterns by classifying in multiresolution subspaces. Our system is able to work with any feature set and any classifier. It consists of multiresolution (MR) decomposition, followed by feature computation and classification in each MR subspace, yielding local decisions that are then combined into a global decision. With 26 texture features alone and a neural network classifier, we obtained an increase in accuracy on the 2D HeLa data set to 95.3%. Conclusion We demonstrate that the space-frequency localized information in the multiresolution subspaces adds significantly to the discriminative power of the system. Moreover, we show that a vastly reduced set of features is sufficient, consisting of our novel modified Haralick texture features. Our proposed system is general, allowing for any combinations of sets of features and any combination of classifiers.