How Many 3D Structures Do We Need to Train a Predictor？

It has been shown that the progress in the determination of membrane protein structure grows exponentially, with approximately the same growth rate as that of the water-soluble proteins. In order to investigate the effect of this, on the performance of prediction algorithms for both α-helical and β-barrel membrane proteins, we conducted a prospective study based on historical records. We trained separate hidden Markov models with different sized training sets and evaluated their performance on topology pred...     

Predicting transmembrane beta-barrels and interstrand residue interactions from sequence

Transmembrane beta-barrel (TMB) proteins are embedded in the outer membrane of Gram-negative bacteria, mitochondria, and chloroplasts. The cellular location and functional diversity of beta-barrel outer membrane proteins (omps) makes them an important protein class. At the present time, very few nonhomologous TMB structures have been determined by X-ray diffraction because of the experimental difficulty encountered in crystallizing transmembrane proteins. A novel method using pairwise interstrand residue statistical potentials derived from globular (nonouter membrane) proteins is introduced to predict the supersecondary structure of transmembrane beta-barrel proteins. The algorithm transFold employs a generalized hidden Markov model (i.e., multitape S-attribute grammar) to describe potential beta-barrel supersecondary structures and then computes by dynamic programming the minimum free energy beta-barrel structure. Hence, the approach can be viewed as a "wrapping" component that may capture folding processes with an initiation stage followed by progressive interaction of the sequence with the already-formed motifs. This approach differs significantly from others, which use traditional machine learning to solve this problem, because it does not require a training phase on known TMB structures and is the first to explicitly capture and predict long-range interactions. TransFold outperforms previous programs for predicting TMBs on smaller (相似文献   

Prediction of transmembrane regions of beta-barrel proteins using ANN- and SVM-based methods

This article describes a method developed for predicting transmembrane beta-barrel regions in membrane proteins using machine learning techniques: artificial neural network (ANN) and support vector machine (SVM). The ANN used in this study is a feed-forward neural network with a standard back-propagation training algorithm. The accuracy of the ANN-based method improved significantly, from 70.4% to 80.5%, when evolutionary information was added to a single sequence as a multiple sequence alignment obtained from PSI-BLAST. We have also developed an SVM-based method using a primary sequence as input and achieved an accuracy of 77.4%. The SVM model was modified by adding 36 physicochemical parameters to the amino acid sequence information. Finally, ANN- and SVM-based methods were combined to utilize the full potential of both techniques. The accuracy and Matthews correlation coefficient (MCC) value of SVM, ANN, and combined method are 78.5%, 80.5%, and 81.8%, and 0.55, 0.63, and 0.64, respectively. These methods were trained and tested on a nonredundant data set of 16 proteins, and performance was evaluated using "leave one out cross-validation" (LOOCV). Based on this study, we have developed a Web server, TBBPred, for predicting transmembrane beta-barrel regions in proteins (available at http://www.imtech.res.in/raghava/tbbpred).     

The beta-barrel finder (BBF) program,allowing identification of outer membrane beta-barrel proteins encoded within prokaryotic genomes

Many outer membrane proteins (OMPs) in Gram-negative bacteria possess known beta-barrel three-dimensional (3D) structures. These proteins, including channel-forming transmembrane porins, are diverse in sequence but exhibit common structural features. We here report computational analyses of six outer membrane proteins of known 3D structures with respect to (1) secondary structure, (2) hydropathy, and (3) amphipathicity. Using these characteristics, as well as the presence of an N-terminal targeting sequence, a program was developed allowing prediction of integral membrane beta-barrel proteins encoded within any completely sequenced prokaryotic genome. This program, termed the beta-barrel finder (BBF) program, was used to analyze the proteins encoded within the Escherichia coli genome. Out of 4290 sequences examined, 118 (2.8%) were retrieved. Of these, almost all known outer membrane proteins with established beta-barrel structures as well as many probable outer membrane proteins were identified. This program should be useful for predicting the occurrence of outer membrane proteins in bacteria with completely sequenced genomes.     

Correct folding of the beta-barrel of the human membrane protein VDAC requires a lipid bilayer

Spontaneous membrane insertion and folding of beta-barrel membrane proteins from an unfolded state into lipid bilayers has been shown previously only for few outer membrane proteins of Gram-negative bacteria. Here we investigated membrane insertion and folding of a human membrane protein, the isoform 1 of the voltage-dependent anion-selective channel (hVDAC1) of mitochondrial outer membranes. Two classes of transmembrane proteins with either alpha-helical or beta-barrel membrane domains are known from the solved high-resolution structures. VDAC forms a transmembrane beta-barrel with an additional N-terminal alpha-helix. We demonstrate that similar to bacterial OmpA, urea-unfolded hVDAC1 spontaneously inserts and folds into lipid bilayers upon denaturant dilution in the absence of folding assistants or energy sources like ATP. Recordings of the voltage-dependence of the single channel conductance confirmed folding of hVDAC1 to its active form. hVDAC1 developed first beta-sheet secondary structure in aqueous solution, while the alpha-helical structure was formed in the presence of lipid or detergent. In stark contrast to bacterial beta-barrel membrane proteins, hVDAC1 formed different structures in detergent micelles and phospholipid bilayers, with higher content of beta-sheet and lower content of alpha-helix when inserted and folded into lipid bilayers. Experiments with mixtures of lipid and detergent indicated that the content of beta-sheet secondary structure in hVDAC1 decreased at increased detergent content. Unlike bacterial beta-barrel membrane proteins, hVDAC1 was not stable even in mild detergents such as LDAO or dodecylmaltoside. Spontaneous folding of outer membrane proteins into lipid bilayers indicates that in cells, the main purpose of membrane-inserted or associated assembly factors may be to select and target beta-barrel membrane proteins towards the outer membrane instead of actively assembling them under consumption of energy as described for the translocons of cytoplasmic membranes.     

Asymmetric amino acid compositions of transmembrane beta-strands

In contrast to water-soluble proteins, membrane proteins reside in a heterogeneous environment, and their surfaces must interact with both polar and apolar membrane regions. As a consequence, the composition of membrane proteins' residues varies substantially between the membrane core and the interfacial regions. The amino acid compositions of helical membrane proteins are also known to be different on the cytoplasmic and extracellular sides of the membrane. Here we report that in the 16 transmembrane beta-barrel structures, the amino acid compositions of lipid-facing residues are different near the N and C termini of the individual strands. Polar amino acids are more prevalent near the C termini than near the N termini, and hydrophobic amino acids show the opposite trend. We suggest that this difference arises because it is easier for polar atoms to escape from the apolar regions of the bilayer at the C terminus of a beta-strand. This new characteristic of beta-barrel membrane proteins enhances our understanding of how a sequence encodes a membrane protein structure and should prove useful in identifying and predicting the structures of trans-membrane beta-barrels.     

Beta barrel trans-membrane proteins: Enhanced prediction using a Bayesian approach

Membrane proteins, which constitute approximately 20% of most genomes, form two main classes: alpha helical and beta barrel transmembrane proteins. Using methods based on Bayesian Networks, a powerful approach for statistical inference, we have sought to address beta-barrel topology prediction. The beta-barrel topology predictor reports individual strand accuracies of 88.6%. The method outlined here represents a potentially important advance in the computational determination of membrane protein topology.     

A functional protein pore with a "retro" transmembrane domain.

Extended retro (reversed) peptide sequences have not previously been accommodated within functional proteins. Here, we show that the entire transmembrane portion of the beta-barrel of the pore-forming protein alpha-hemolysin can be formed by retrosequences comprising a total of 175 amino acid residues, 25 contributed by the central sequence of each subunit of the heptameric pore. The properties of wild-type and retro heptamers in planar bilayers are similar. The single-channel conductance of the retro pore is 15% less than that of the wild-type heptamer and its current-voltage relationship denotes close to ohmic behavior, while the wild-type pore is weakly rectifying. Both wild-type and retro pores are very weakly anion selective. These results and the examination of molecular models suggest that beta-barrels may be especially accepting of retro sequences compared to other protein folds. Indeed, the ability to form a retro domain could be diagnostic of a beta-barrel, explaining, for example, the activity of the retro forms of many membrane-permeabilizing peptides. By contrast with the wild-type subunits, monomeric retro subunits undergo premature assembly in the absence of membranes, most likely because the altered central sequence fails to interact with the remainder of the subunit, thereby initiating assembly. Despite this difficulty, a technique was devised for obtaining heteromeric pores containing both wild-type and retro subunits. Most probably as a consequence of unfavorable interstrand side-chain interactions, the heteromeric pores are less stable than either the wild-type or retro homoheptamers, as judged by the presence of subconductance states in single-channel recordings. Knowledge about the extraordinary plasticity of the transmembrane beta-barrel of alpha-hemolysin will be very useful in the de novo design of functional membrane proteins based on the beta-barrel motif.     

Gibbs motif sampling: detection of bacterial outer membrane protein repeats.

The detection and alignment of locally conserved regions (motifs) in multiple sequences can provide insight into protein structure, function, and evolution. A new Gibbs sampling algorithm is described that detects motif-encoding regions in sequences and optimally partitions them into distinct motif models; this is illustrated using a set of immunoglobulin fold proteins. When applied to sequences sharing a single motif, the sampler can be used to classify motif regions into related submodels, as is illustrated using helix-turn-helix DNA-binding proteins. Other statistically based procedures are described for searching a database for sequences matching motifs found by the sampler. When applied to a set of 32 very distantly related bacterial integral outer membrane proteins, the sampler revealed that they share a subtle, repetitive motif. Although BLAST (Altschul SF et al., 1990, J Mol Biol 215:403-410) fails to detect significant pairwise similarity between any of the sequences, the repeats present in these outer membrane proteins, taken as a whole, are highly significant (based on a generally applicable statistical test for motifs described here). Analysis of bacterial porins with known trimeric beta-barrel structure and related proteins reveals a similar repetitive motif corresponding to alternating membrane-spanning beta-strands. These beta-strands occur on the membrane interface (as opposed to the trimeric interface) of the beta-barrel. The broad conservation and structural location of these repeats suggests that they play important functional roles.     

Topology of the outer membrane usher PapC determined by site-directed fluorescence labeling

In contrast to typical membrane proteins that span the lipid bilayer via transmembrane alpha-helices, bacterial outer membrane proteins adopt a beta-barrel architecture composed of antiparallel transmembrane beta-strands. The topology of outer membrane proteins is difficult to predict accurately using computer algorithms, and topology mapping protocols commonly used for alpha-helical membrane proteins do not work for beta-barrel proteins. We present here the topology of the PapC usher, an outer membrane protein required for assembly and secretion of P pili by the chaperone/usher pathway in uropathogenic Escherichia coli. An initial attempt to map PapC topology by insertion of protease cleavage sites was largely unsuccessful due to lack of cleavage at most sites and the requirement to disrupt the outer membrane to identify periplasmic sites. We therefore adapted a site-directed fluorescence labeling technique to permit topology mapping of outer membrane proteins using small molecule probes in intact bacteria. Using this method, we demonstrated that PapC has the potential to encode up to 32 transmembrane beta-strands. Based on experimental evidence, we propose that the usher consists of an N-terminal beta-barrel domain comprised of 26 beta-strands and that a distinct C-terminal domain is not inserted into the membrane but is located instead within the lumen of the N-terminal beta-barrel similar to the plug domains encoded by the outer membrane iron-siderophore uptake proteins.     

A consensus algorithm to screen genomes for novel families of transmembrane beta barrel proteins

The ability to search sequence datasets for membrane spanning proteins is an important requirement for genome annotation. However, the development of algorithms to identify novel types of transmembrane beta-barrel (TMB) protein has proven substantially harder than for transmembrane helical proteins, owing to a shorter TM domain in which only alternate residues are hydrophobic. Although recent reports have described important improvements in the development of such algorithms, there is still concern over their ability to confidently screen genomes. Here we describe a new algorithm combining composition and hidden Markov model topology based classifiers (called TMB-Hunt2), which achieves a crossvalidation accuracy of >95%, with 96.7% precision and 94.2% recall. An overview is given of the algorithm design, with a thorough assessment of performance and application to a number of genomes. Of particular note is that TMB/extracellular protein discrimination is significantly more difficult than TMB/cytoplasmic protein discrimination, with the predictor correctly rejecting just 74% of extracellular proteins, in comparison to 98% of cytoplasmic proteins. Focus is given to directions for further improvements in TMB/non-TMB protein discrimination, with a call for the development of standardized tests and assessments of such algorithms. Tools and datasets are made available through a website called TMB-Web (http://www.bioinformatics.leeds.ac.uk/TMB-Web/TMB-Hunt2).     

A predictor of transmembrane α-helix domains of proteins based on neural networks

Back-propagation, feed-forward neural networks are used to predict a-helical transmembrane segments of proteins. The networks are trained on the few membrane proteins whose transmembrane -helix domains are known to atomic or nearly atomic resolution. When testing is performed with a jackknife procedure on the proteins of the training set, the fraction of total correct assignments is as high as 0.87, with an average length for the transmembrane segments of 20 residues. The method correctly fails to predict any transmembrane domain for porin, whose transmembrane segments are -sheets. When tested on globular proteins, lower and upper limits of 1.6 and 3.5% for a total of 26826 residues are determined for the mispredicted cases, indicating that the predictor is highly specific for -helical domains of membrane proteins. The predictor is also tested on 37 membrane proteins whose transmembrane topology is partially known. The overall accuracy is 0.90, two percentage points higher than that obtained with statistical methods. The reliability of the prediction is 100% for 60% of the total 18242 predicted residues of membrane proteins. Our results show that the local directional information automatically extracted by the neural networks during the training phase plays a key role in determining the accuracy of the prediction. Correspondence to: R. Casadio     

Toward genomic identification of β-barrel membrane proteins: Composition and architecture of known structures

The amino acid composition and architecture of all beta-barrel membrane proteins of known three-dimensional structure have been examined to generate information that will be useful in identifying beta-barrels in genome databases. The database consists of 15 nonredundant structures, including several novel, recent structures. Known structures include monomeric, dimeric, and trimeric beta-barrels with between 8 and 22 membrane-spanning beta-strands each. For this analysis the membrane-interacting surfaces of the beta-barrels were identified with an experimentally derived, whole-residue hydrophobicity scale, and then the barrels were aligned normal to the bilayer and the position of the bilayer midplane was determined for each protein from the hydrophobicity profile. The abundance of each amino acid, relative to the genomic abundance, was calculated for the barrel exterior and interior. The architecture and diversity of known beta-barrels was also examined. For example, the distribution of rise-per-residue values perpendicular to the bilayer plane was found to be 2.7 +/- 0.25 A per residue, or about 10 +/- 1 residues across the membrane. Also, as noted by other authors, nearly every known membrane-spanning beta-barrel strand was found to have a short loop of seven residues or less connecting it to at least one adjacent strand. Using this information we have begun to generate rapid screening algorithms for the identification of beta-barrel membrane proteins in genomic databases. Application of one algorithm to the genomes of Escherichia coli and Pseudomonas aeruginosa confirms its ability to identify beta-barrels, and reveals dozens of unidentified open reading frames that potentially code for beta-barrel outer membrane proteins.     

A simple method for predicting transmembrane alpha helices with better accuracy.

The prediction of a protein's structure from its amino acid sequence has been a long-standing goal of molecular biology. In this work, a new set of conformational parameters for membrane spanning alpha helices was developed using the information from the topology of 70 membrane proteins. Based on these conformational parameters, a simple algorithm has been formulated to predict the transmembrane alpha helices in membrane proteins. A FORTRAN program has been developed which takes the amino acid sequence as input and gives the predicted transmembrane alpha-helices as output. The present method correctly identifies 295 transmembrane helical segments in 70 membrane proteins with only two overpredictions. Furthermore, this method predicts all 45 transmembrane helices in the photosynthetic reaction center, bacteriorhodopsin and cytochrome c oxidase to an 86% level of accuracy and so is better than all other methods published to date.     

Predicting transmembrane beta-barrels in proteomes

Very few methods address the problem of predicting beta-barrel membrane proteins directly from sequence. One reason is that only very few high-resolution structures for transmembrane beta-barrel (TMB) proteins have been determined thus far. Here we introduced the design, statistics and results of a novel profile-based hidden Markov model for the prediction and discrimination of TMBs. The method carefully attempts to avoid over-fitting the sparse experimental data. While our model training and scoring procedures were very similar to a recently published work, the architecture and structure-based labelling were significantly different. In particular, we introduced a new definition of beta- hairpin motifs, explicit state modelling of transmembrane strands, and a log-odds whole-protein discrimination score. The resulting method reached an overall four-state (up-, down-strand, periplasmic-, outer-loop) accuracy as high as 86%. Furthermore, accurately discriminated TMB from non-TMB proteins (45% coverage at 100% accuracy). This high precision enabled the application to 72 entirely sequenced Gram-negative bacteria. We found over 164 previously uncharacterized TMB proteins at high confidence. Database searches did not implicate any of these proteins with membranes. We challenge that the vast majority of our 164 predictions will eventually be verified experimentally. All proteome predictions and the PROFtmb prediction method are available at http://www.rostlab.org/ services/PROFtmb/.     

A sequence-profile-based HMM for predicting and discriminating beta barrel membrane proteins

MOTIVATION: Membrane proteins are an abundant and functionally relevant subset of proteins that putatively include from about 15 up to 30% of the proteome of organisms fully sequenced. These estimates are mainly computed on the basis of sequence comparison and membrane protein prediction. It is therefore urgent to develop methods capable of selecting membrane proteins especially in the case of outer membrane proteins, barely taken into consideration when proteome wide analysis is performed. This will also help protein annotation when no homologous sequence is found in the database. Outer membrane proteins solved so far at atomic resolution interact with the external membrane of bacteria with a characteristic beta barrel structure comprising different even numbers of beta strands (beta barrel membrane proteins). In this they differ from the membrane proteins of the cytoplasmic membrane endowed with alpha helix bundles (all alpha membrane proteins) and need specialised predictors. RESULTS: We develop a HMM model, which can predict the topology of beta barrel membrane proteins using, as input, evolutionary information. The model is cyclic with 6 types of states: two for the beta strand transmembrane core, one for the beta strand cap on either side of the membrane, one for the inner loop, one for the outer loop and one for the globular domain state in the middle of each loop. The development of a specific input for HMM based on multiple sequence alignment is novel. The accuracy per residue of the model is 83% when a jack knife procedure is adopted. With a model optimisation method using a dynamic programming algorithm seven topological models out of the twelve proteins included in the testing set are also correctly predicted. When used as a discriminator, the model is rather selective. At a fixed probability value, it retains 84% of a non-redundant set comprising 145 sequences of well-annotated outer membrane proteins. Concomitantly, it correctly rejects 90% of a set of globular proteins including about 1200 chains with low sequence identity (<30%) and 90% of a set of all alpha membrane proteins, including 188 chains.     

Discrimination of outer membrane proteins by a new measure of information discrepancy

Discriminating outer membrane proteins for globular proteins (GPs) and other types of membrane proteins from genomic sequences is an important and hot topic. In this paper, a measure based on information discrepancy is proposed and applied to the discrimination of outer membrane proteins. It differs from previous methods which are based on amino acid composition. Our approach focuses on the comparison of subsequence distributions and takes into account the effect of residue order in protein primary structures. As a result, the new approach outperforms all previous methods on the same benchmark datasets. In particular, we show that the proposed approach has correctly identified the outer membrane proteins at an accuracy of 99% for the training set of 337 proteins and has correctly excluded the GPs at an accuracy of 86% in a non-redundant dataset of 668 proteins. Furthermore, this method is able to correctly exclude alpha-helical membrane proteins at an accuracy of 100%.     

Transmembrane helices predicted at 95% accuracy.

We describe a neural network system that predicts the locations of transmembrane helices in integral membrane proteins. By using evolutionary information as input to the network system, the method significantly improved on a previously published neural network prediction method that had been based on single sequence information. The input data were derived from multiple alignments for each position in a window of 13 adjacent residues: amino acid frequency, conservation weights, number of insertions and deletions, and position of the window with respect to the ends of the protein chain. Additional input was the amino acid composition and length of the whole protein. A rigorous cross-validation test on 69 proteins with experimentally determined locations of transmembrane segments yielded an overall two-state per-residue accuracy of 95%. About 94% of all segments were predicted correctly. When applied to known globular proteins as a negative control, the network system incorrectly predicted fewer than 5% of globular proteins as having transmembrane helices. The method was applied to all 269 open reading frames from the complete yeast VIII chromosome. For 59 of these, at least two transmembrane helices were predicted. Thus, the prediction is that about one-fourth of all proteins from yeast VIII contain one transmembrane helix, and some 20%, more than one.     

Sequence motifs and antimotifs in beta-barrel membrane proteins from a genome-wide analysis: the Ala-Tyr dichotomy and chaperone binding motifs

Beta-barrel membrane proteins are found in the outer membrane of gram-negative bacteria, mitochondria, and chloroplasts. Although sequence motifs have been studied in alpha-helical membrane proteins and have been shown to play important roles in their assembly, it is not clear whether over-represented motifs and under-represented anti-motifs exist in beta-barrel membrane proteins. We have developed probabilistic models to identify sequence motifs of residue pairs on the same strand separated by an arbitrary number of residues. A rigorous statistical model is essential for this study because of the difficulty associated with the short length of the strands and the small amount of structural data. By comparing to the null model of exhaustive permutation of residues within the same beta-strand, propensity values of sequence patterns of two residues and p-values measuring statistical significance are calculated exactly by several analytical formulae we have developed or by enumeration. We find that there are characteristic sequence motifs and antimotifs in transmembrane (TM) beta-strands. The amino acid Tyr plays an important role in several such motifs. We find a general dichotomy consisting of favorable Aliphatic-Tyr sequence motifs and unfavorable Tyr-Aliphatic antimotifs. Tyr is also part of a terminal motif, YxF, which is likely to be important for chaperone binding. Our results also suggest several experiments that can help to elucidate the mechanisms of in vitro and in vivo folding of beta-barrel membrane proteins.     

Predicting the solvent accessibility of transmembrane residues from protein sequence

In this study, we propose a novel method to predict the solvent accessible surface areas of transmembrane residues. For both transmembrane alpha-helix and beta-barrel residues, the correlation coefficients between the predicted and observed accessible surface areas are around 0.65. On the basis of predicted accessible surface areas, residues exposed to the lipid environment or buried inside a protein can be identified by using certain cutoff thresholds. We have extensively examined our approach based on different definitions of accessible surface areas and a variety of sets of control parameters. Given that experimentally determining the structures of membrane proteins is very difficult and membrane proteins are actually abundant in nature, our approach is useful for theoretically modeling membrane protein tertiary structures, particularly for modeling the assembly of transmembrane domains. This approach can be used to annotate the membrane proteins in proteomes to provide extra structural and functional information.