Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes

We describe and validate a new membrane protein topology prediction method, TMHMM, based on a hidden Markov model. We present a detailed analysis of TMHMM's performance, and show that it correctly predicts 97-98 % of the transmembrane helices. Additionally, TMHMM can discriminate between soluble and membrane proteins with both specificity and sensitivity better than 99 %, although the accuracy drops when signal peptides are present. This high degree of accuracy allowed us to predict reliably integral membrane proteins in a large collection of genomes. Based on these predictions, we estimate that 20-30 % of all genes in most genomes encode membrane proteins, which is in agreement with previous estimates. We further discovered that proteins with N(in)-C(in) topologies are strongly preferred in all examined organisms, except Caenorhabditis elegans, where the large number of 7TM receptors increases the counts for N(out)-C(in) topologies. We discuss the possible relevance of this finding for our understanding of membrane protein assembly mechanisms. A TMHMM prediction service is available at http://www.cbs.dtu.dk/services/TMHMM/.     

An improved hidden Markov model for transmembrane protein detection and topology prediction and its applications to complete genomes

MOTIVATION: Knowledge of the transmembrane helical topology can help identify binding sites and infer functions for membrane proteins. However, because membrane proteins are hard to solubilize and purify, only a very small amount of membrane proteins have structure and topology experimentally determined. This has motivated various computational methods for predicting the topology of membrane proteins. RESULTS: We present an improved hidden Markov model, TMMOD, for the identification and topology prediction of transmembrane proteins. Our model uses TMHMM as a prototype, but differs from TMHMM by the architecture of the submodels for loops on both sides of the membrane and also by the model training procedure. In cross-validation experiments using a set of 83 transmembrane proteins with known topology, TMMOD outperformed TMHMM and other existing methods, with an accuracy of 89% for both topology and locations. In another experiment using a separate set of 160 transmembrane proteins, TMMOD had 84% for topology and 89% for locations. When utilized for identifying transmembrane proteins from non-transmembrane proteins, particularly signal peptides, TMMOD has consistently fewer false positives than TMHMM does. Application of TMMOD to a collection of complete genomes shows that the number of predicted membrane proteins accounts for approximately 20-30% of all genes in those genomes, and that the topology where both the N- and C-termini are in the cytoplasm is dominant in these organisms except for Caenorhabditis elegans. AVAILABILITY: http://liao.cis.udel.edu/website/servers/TMMOD/     

The use of functional domains to improve transmembrane protein topology prediction

Transmembrane proteins affect vital cellular functions and pathogenesis, and are a focus of drug design. It is difficult to obtain diffraction quality crystals to study transmembrane protein structure. Computational tools for transmembrane protein topology prediction fill in the gap between the abundance of transmembrane proteins and the scarcity of known membrane protein structures. Their prediction accuracy is still inadequate: TMHMM, the current state-of-the-art method, has less than 52% accuracy in topology prediction on one set of transmembrane proteins of known topology. Based on the observation that there are functional domains that occur preferentially internal or external to the membrane, we have extended the model of TMHMM to incorporate functional domains, using a probabilistic approach originally developed for computational gene finding. Our extension is better than TMHMM in predicting the topology of transmembrane proteins. As prediction of functional domain improves, our system's prediction accuracy will likely improve as well.     

Prediction of partial membrane protein topologies using a consensus approach

We have developed a method to reliably identify partial membrane protein topologies using the consensus of five topology prediction methods. When evaluated on a test set of experimentally characterized proteins, we find that approximately 90% of the partial consensus topologies are correctly predicted in membrane proteins from prokaryotic as well as eukaryotic organisms. Whole-genome analysis reveals that a reliable partial consensus topology can be predicted for approximately 70% of all membrane proteins in a typical bacterial genome and for approximately 55% of all membrane proteins in a typical eukaryotic genome. The average fraction of sequence length covered by a partial consensus topology is 44% for the prokaryotic proteins and 17% for the eukaryotic proteins in our test set, and similar numbers are found when the algorithm is applied to whole genomes. Reliably predicted partial topologies may simplify experimental determinations of membrane protein topology.     

Reliability of transmembrane predictions in whole-genome data

Transmembrane prediction methods are generally benchmarked on a set of proteins with experimentally verified topology. We have investigated if the accuracy measured on such datasets can be expected in an unbiased genomic analysis, or if there is a bias towards 'easily predictable' proteins in the benchmark datasets. As a measurement of accuracy, the concordance of the results from five different prediction methods was used (TMHMM, PHD, HMMTOP, MEMSAT, and TOPPRED). The benchmark dataset showed significantly higher levels (up to five times) of agreement between different methods than in 10 tested genomes. We have also analyzed which programs are most prone to make mispredictions by measuring the frequency of one-out-of-five disagreeing predictions.     

Evaluation of methods for the prediction of membrane spanning regions.

MOTIVATION: A variety of tools are available to predict the topology of transmembrane proteins. To date no independent evaluation of the performance of these tools has been published. A better understanding of the strengths and weaknesses of the different tools would guide both the biologist and the bioinformatician to make better predictions of membrane protein topology. RESULTS: Here we present an evaluation of the performance of the currently best known and most widely used methods for the prediction of transmembrane regions in proteins. Our results show that TMHMM is currently the best performing transmembrane prediction program.     

Consensus predictions of membrane protein topology

We have explored the possibility that consensus predictions of membrane protein topology might provide a means to estimate the reliability of a predicted topology. Using five current topology prediction methods and a test set of 60 Escherichia coli inner membrane proteins with experimentally determined topologies, we find that prediction performance varies strongly with the number of methods that agree, and that the topology of nearly half of all E. coli inner membrane proteins can be predicted with high reliability (>90% correct predictions) by a simple majority-vote approach.     

Improved membrane protein topology prediction by domain assignments

Topology predictions for integral membrane proteins can be substantially improved if parts of the protein can be constrained to a given in/out location relative to the membrane using experimental data or other information. Here, we have identified a set of 367 domains in the SMART database that, when found in soluble proteins, have compartment-specific localization of a kind relevant for membrane protein topology prediction. Using these domains as prediction constraints, we are able to provide high-quality topology models for 11% of the membrane proteins extracted from 38 eukaryotic genomes. Two-thirds of these proteins are single spanning, a group of proteins for which current topology prediction methods perform particularly poorly.     

Rapid screening of membrane topology of secondary transport proteins

Limited experimental data may be very useful to discriminate between membrane topology models of membrane proteins derived from different methods. A membrane topology screening method is proposed by which the cellular disposition of three positions in a membrane protein are determined, the N- and the C-termini and a position in the middle of the protein. The method involves amplification of the encoding genes or gene fragments by PCR, rapid cloning in dedicated vectors by ligation independent cloning, and determination of the cellular disposition of the three sites using conventional techniques. The N-terminus was determined by labeling with a fluorescent probe, the central position and the C-terminus by the reporter fusion technique using alkaline phosphatase (PhoA) and green fluorescence protein (GFP) as reporters. The method was evaluated using 16 transporter proteins of known function from four different structural classes. For 13 proteins a complete set of three localizations was obtained. The experimental data was used to discriminate between membrane topology models predicted by TMHMM, a widely used predictor using the amino acid sequence as input and by MemGen that uses hydropathy profile alignment and known 3D structures or existing models. It follows that in those cases where the models from the two methods were similar, the models were consistent with the experimental data. In those cases where the models differed, the MemGen model agreed with the experimental data. Three more recent predictors, MEMSAT3, OCTOPUS and TOPCONS showed a significantly higher consistency with the experimental data than observed with TMHMM.     

A guideline to proteome-wide α-helical membrane protein topology predictions

For current state-of-the-art methods, the prediction of correct topology of membrane proteins has been reported to be above 80%. However, this performance has only been observed in small and possibly biased data sets obtained from protein structures or biochemical assays. Here, we test a number of topology predictors on an "unseen" set of proteins of known structure and also on four "genome-scale" data sets, including one recent large set of experimentally validated human membrane proteins with glycosylated sites. The set of glycosylated proteins is also used to examine the ability of prediction methods to separate membrane from nonmembrane proteins. The results show that methods utilizing multiple sequence alignments are overall superior to methods that do not. The best performance is obtained by TOPCONS, a consensus method that combines several of the other prediction methods. The best methods to distinguish membrane from nonmembrane proteins belong to the "Phobius" group of predictors. We further observe that the reported high accuracies in the smaller benchmark sets are not quite maintained in larger scale benchmarks. Instead, we estimate the performance of the best prediction methods for eukaryotic membrane proteins to be between 60% and 70%. The low agreement between predictions from different methods questions earlier estimates about the global properties of the membrane proteome. Finally, we suggest a pipeline to estimate these properties using a combination of the best predictors that could be applied in large-scale proteomics studies of membrane proteins.     

Enhanced membrane protein topology prediction using a hierarchical classification method and a new scoring function

The prediction of transmembrane (TM) helix and topology provides important information about the structure and function of a membrane protein. Due to the experimental difficulties in obtaining a high-resolution model, computational methods are highly desirable. In this paper, we present a hierarchical classification method using support vector machines (SVMs) that integrates selected features by capturing the sequence-to-structure relationship and developing a new scoring function based on membrane protein folding. The proposed approach is evaluated on low- and high-resolution data sets with cross-validation, and the topology (sidedness) prediction accuracy reaches as high as 90%. Our method is also found to correctly predict both the location of TM helices and the topology for 69% of the low-resolution benchmark set. We also test our method for discrimination between soluble and membrane proteins and achieve very low overall false positive (0.5%) and false negative rates (0 to approximately 1.2%). Lastly, the analysis of the scoring function suggests that the topogeneses of single-spanning and multispanning TM proteins have different levels of complexity, and the consideration of interloop topogenic interactions for the latter is the key to achieving better predictions. This method can facilitate the annotation of membrane proteomes to extract useful structural and functional information. It is publicly available at http://bio-cluster.iis.sinica.edu.tw/~bioapp/SVMtop.     

Rapid membrane protein topology prediction

State-of-the-art methods for topology of α-helical membrane proteins are based on the use of time-consuming multiple sequence alignments obtained from PSI-BLAST or other sources. Here, we examine if it is possible to use the consensus of topology prediction methods that are based on single sequences to obtain a similar accuracy as the more accurate multiple sequence-based methods. Here, we show that TOPCONS-single performs better than any of the other topology prediction methods tested here, but ~6% worse than the best method that is utilizing multiple sequence alignments. AVAILABILITY AND IMPLEMENTATION: TOPCONS-single is available as a web server from http://single.topcons.net/ and is also included for local installation from the web site. In addition, consensus-based topology predictions for the entire international protein index (IPI) is available from the web server and will be updated at regular intervals.     

Topology prediction for helical transmembrane proteins at 86% accuracy.

Previously, we introduced a neural network system predicting locations of transmembrane helices (HTMs) based on evolutionary profiles (PHDhtm, Rost B, Casadio R, Fariselli P, Sander C, 1995, Protein Sci 4:521-533). Here, we describe an improvement and an extension of that system. The improvement is achieved by a dynamic programming-like algorithm that optimizes helices compatible with the neural network output. The extension is the prediction of topology (orientation of first loop region with respect to membrane) by applying to the refined prediction the observation that positively charged residues are more abundant in extra-cytoplasmic regions. Furthermore, we introduce a method to reduce the number of false positives, i.e., proteins falsely predicted with membrane helices. The evaluation of prediction accuracy is based on a cross-validation and a double-blind test set (in total 131 proteins). The final method appears to be more accurate than other methods published: (1) For almost 89% (+/-3%) of the test proteins, all HTMs are predicted correctly. (2) For more than 86% (+/-3%) of the proteins, topology is predicted correctly. (3) We define reliability indices that correlate with prediction accuracy: for one half of the proteins, segment accuracy raises to 98%; and for two-thirds, accuracy of topology prediction is 95%. (4) The rate of proteins for which HTMs are predicted falsely is below 2% (+/-1%). Finally, the method is applied to 1,616 sequences of Haemophilus influenzae. We predict 19% of the genome sequences to contain one or more HTMs. This appears to be lower than what we predicted previously for the yeast VIII chromosome (about 25%).     

Estimating the length of transmembrane helices using Z-coordinate predictions

Zpred2 is an improved version of ZPRED, a predictor for the Z-coordinates of alpha-helical membrane proteins, that is, the distance of the residues from the center of the membrane. Using principal component analysis and a set of neural networks, Zpred2 analyzes data extracted from the amino acid sequence, the predicted topology, and evolutionary profiles. Zpred2 achieves an average accuracy error of 2.18 A (2.17 A when an independent test set is used), an improvement by 15% compared to the previous version. We show that this accuracy is sufficient to enable the predictions of helix lengths with a correlation coefficient of 0.41. As a comparison, two state-of-the-art HMM-based topology prediction methods manage to predict the helix lengths with a correlation coefficient of less than 0.1. In addition, we applied Zpred2 to two other problems, the re-entrant region identification and model validation. Re-entrants were able to be detected with a certain consistency, but not better than with previous approaches, while incorrect models as well as mispredicted helices of transmembrane proteins could be distinguished based on the Z-coordinate predictions.     

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

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

Folding in lipid membranes (FILM): a novel method for the prediction of small membrane protein 3D structures

We present the results of applying a novel knowledge-based method (FILM) to the prediction of small membrane protein structures. The basis of the method is the addition of a membrane potential to the energy terms (pairwise, solvation, steric, and hydrogen bonding) of a previously developed ab initio technique for the prediction of tertiary structure of globular proteins (FRAGFOLD). The method is based on the assembly of supersecondary structural fragments taken from a library of highly resolved protein structures using a standard simulated annealing algorithm. The membrane potential has been derived by the statistical analysis of a data set made of 640 transmembrane helices with experimentally defined topology and belonging to 133 proteins extracted from the SWISS-PROT database. Results obtained by applying the method to small membrane proteins of known 3D structure show that the method is able to predict, at a reasonable accuracy level, both the helix topology and the conformations of these proteins.     

TMBHMM: a frequency profile based HMM for predicting the topology of transmembrane beta barrel proteins and the exposure status of transmembrane residues

Transmembrane beta barrel (TMB) proteins are found in the outer membranes of bacteria, mitochondria and chloroplasts. TMBs are involved in a variety of functions such as mediating flux of metabolites and active transport of siderophores, enzymes and structural proteins, and in the translocation across or insertion into membranes. We present here TMBHMM, a computational method based on a hidden Markov model for predicting the structural topology of putative TMBs from sequence. In addition to predicting transmembrane strands, TMBHMM also predicts the exposure status (i.e., exposed to the membrane or hidden in the protein structure) of the residues in the transmembrane region, which is a novel feature of the TMBHMM method. Furthermore, TMBHMM can also predict the membrane residues that are not part of beta barrel forming strands. The training of the TMBHMM was performed on a non-redundant data set of 19 TMBs. The self-consistency test yielded Q(2) accuracy of 0.87, Q(3) accuracy of 0.83, Matthews correlation coefficient of 0.74 and SOV for beta strand of 0.95. In this self-consistency test the method predicted 83% of transmembrane residues with correct exposure status. On an unseen, non-redundant test data set of 10 proteins, the 2-state and 3-state TMBHMM prediction accuracies are around 73% and 72%, respectively, and are comparable to other methods from the literature. The TMBHMM web server takes an amino acid sequence or a multiple sequence alignment as an input and predicts the exposure status and the structural topology as output. The TMBHMM web server is available under the tmbhmm tab at: http://service.bioinformatik.uni-saarland.de/tmx-site/.     

Estimation of membrane proteins in the human proteome

Genomics and proteomics have added valuable information to our knowledgebase of the human biological system including the discovery of therapeutic targets and disease biomarkers. However, molecular profiling studies commonly result in the identification of novel proteins of unknown localization. A class of proteins of special interest is membrane proteins, in particular plasma membrane proteins. Despite their biological and medical significance, the 3-dimensional structures of less than 1% of plasma membrane proteins have been determined. In order to aid in identification of membrane proteins, a number of computational methods have been developed. These tools operate by predicting the presence of transmembrane segments. Here, we utilized five topology prediction methods (TMHMM, SOSUI, waveTM, HMMTOP, and TopPred II) in order to estimate the ratio of integral membrane proteins in the human proteome. These methods employ different algorithms and include a newly-developed method (waveTM) that has yet to be tested on a large proteome database. Since these tools are prone for error mainly as a result of falsely predicting signal peptides as transmembrane segments, we have utilized an additional method, SignalP. Based on our analyses, the ratio of human proteins with transmembrane segments is estimated to fall between 15% and 39% with a consensus of 13%. Agreement among the programs is reduced further when both a positive identification of a membrane protein and the number of transmembrane segments per protein are considered. Such a broad range of prediction depends on the selectivity of the individual method in predicting integral membrane proteins. These methods can play a critical role in determining protein structure and, hence, identifying suitable drug targets in humans.     

α-螺旋跨膜蛋白拓扑结构预测方法的评价

在基因组数据中,有20%～30%的产物被预测为跨膜蛋白,本文通过对膜蛋白拓扑结构预测方法进行分析,并评价其结果,为选择更合适的拓扑结构预测方法预测膜蛋白结构。通过对目前已有的拓扑结构预测方法的评价分析,可以为我们在实际工作中提供重要的参考。比如对一个未知拓扑结构的跨膜蛋白序列,我们可以先进行是否含有信号肽的预测,参考Polyphobius和SignalP两种方法,若两种方法预测结果不一致,综合上述对两种方法的评价,Polyphobius预测的综合能力较好,可取其预测的结果,一旦确定含有信号肽,则N端必然位于膜外侧。然后结合序列的长度,判断蛋白是单跨膜还是多重跨膜,即可参照上述评价结果,选择合适的拓扑结构预测方法进行预测。     

Secreted protein prediction system combining CJ-SPHMM,TMHMM, and PSORT

To increase the coverage of secreted protein prediction, we describe a combination strategy. Instead of using a single method, we combine Hidden Markov Model (HMM)-based methods CJ-SPHMM and TMHMM with PSORT in secreted protein prediction. CJ-SPHMM is an HMM-based signal peptide prediction method, while TMHMM is an HMM-based transmembrane (TM) protein prediction algorithm. With CJ-SPHMM and TMHMM, proteins with predicted signal peptide and without predicted TM regions are taken as putative secreted proteins. This HMM-based approach predicts secreted protein with Ac (Accuracy) at 0.82 and Cc (Correlation coefficient) at 0.75, which are similar to PSORT with Ac at 0.82 and Cc at 0.76. When we further complement the HMM-based method, i.e., CJ-SPHMM + TMHMM with PSORT in secreted protein prediction, the Ac value is increased to 0.86 and the Cc value is increased to 0.81. Taking this combination strategy to search putative secreted proteins from the International Protein Index (IPI) maintained at the European Bioinformatics Institute (EBI), we constructed a putative human secretome with 5235 proteins. The prediction system described here can also be applied to predicting secreted proteins from other vertebrate proteomes. Availability: The CJ-SPHMM and predicted secreted proteins are available at: ftp://ftp.cbi.pku.edu.cn/pub/secreted-protein/