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%).     

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.     

Prediction of helix-helix contacts and interacting helices in polytopic membrane proteins using neural networks

Despite rapidly increasing numbers of available 3D structures, membrane proteins still account for less than 1% of all structures in the Protein Data Bank. Recent high-resolution structures indicate a clearly broader structural diversity of membrane proteins than initially anticipated, motivating the development of reliable structure prediction methods specifically tailored for this class of molecules. One important prediction target capturing all major aspects of a protein's 3D structure is its contact map. Our analysis shows that computational methods trained to predict residue contacts in globular proteins perform poorly when applied to membrane proteins. We have recently published a method to identify interacting alpha-helices in membrane proteins based on the analysis of coevolving residues in predicted transmembrane regions. Here, we present a substantially improved algorithm for the same problem, which uses a newly developed neural network approach to predict helix-helix contacts. In addition to the input features commonly used for contact prediction of soluble proteins, such as windowed residue profiles and residue distance in the sequence, our network also incorporates features that apply to membrane proteins only, such as residue position within the transmembrane segment and its orientation toward the lipophilic environment. The obtained neural network can predict contacts between residues in transmembrane segments with nearly 26% accuracy. It is therefore the first published contact predictor developed specifically for membrane proteins performing with equal accuracy to state-of-the-art contact predictors available for soluble proteins. The predicted helix-helix contacts were employed in a second step to identify interacting helices. For our dataset consisting of 62 membrane proteins of solved structure, we gained an accuracy of 78.1%. Because the reliable prediction of helix interaction patterns is an important step in the classification and prediction of membrane protein folds, our method will be a helpful tool in compiling a structural census of membrane proteins.     

Proportion of membrane proteins in proteomes of 15 single-cell organisms analyzed by the SOSUI prediction system

A software system, SOSUI, was previously developed for discriminating between soluble and membrane proteins and predicting transmembrane regions (Hirokawa et al., Bioinformatics, 14 (1998) 378-379). The performance of the system was 99% for the discrimination between two types of proteins and 96% for the prediction of transmembrane helices. When all of the amino acid sequences from 15 single-cell organisms were analyzed by SOSUI, the proportion of predicted polytopic membrane proteins showed an almost constant value of 15-20%, irrespective of the total genome size. However, single-cell organisms appeared to be categorized in terms of the preference of the number of transmembrane segments: species with small genomes were characterized by a significant peak at a helix number of approximately six or seven; species with large genomes showed a peak at 10 or 11 helices; and species with intermediate genome sizes showed a monotonous decrease of the population of membrane proteins against the number of transmembrane helices.     

Role of evolutionary information in prediction of aromatic-backbone NH interactions in proteins

In this study, an attempt has been made to develop a neural network-based method for predicting segments in proteins containing aromatic-backbone NH (Ar-NH) interactions using multiple sequence alignment. We have analyzed 3121 segments seven residues long containing Ar-NH interactions, extracted from 2298 non-redundant protein structures where no two proteins have more than 25% sequence identity. Two consecutive feed-forward neural networks with a single hidden layer have been trained with standard back-propagation as learning algorithm. The performance of the method improves from 0.12 to 0.15 in terms of Matthews correlation coefficient (MCC) value when evolutionary information (multiple alignment obtained from PSI-BLAST) is used as input instead of a single sequence. The performance of the method further improves from MCC 0.15 to 0.20 when secondary structure information predicted by PSIPRED is incorporated in the prediction. The final network yields an overall prediction accuracy of 70.1% and an MCC of 0.20 when tested by five-fold cross-validation. Overall the performance is 15.2% higher than the random prediction. The method consists of two neural networks: (i) a sequence-to-structure network which predicts the aromatic residues involved in Ar-NH interaction from multiple alignment of protein sequences and (ii) a structure-to structure network where the input consists of the output obtained from the first network and predicted secondary structure. Further, the actual position of the donor residue within the 'potential' predicted fragment has been predicted using a separate sequence-to-structure neural network. Based on the present study, a server Ar_NHPred has been developed which predicts Ar-NH interaction in a given amino acid sequence. The web server Ar_NHPred is available at and (mirror site).     

Computational differentiation of N-terminal signal peptides and transmembrane helices

Signal peptides and transmembrane helices both contain a stretch of hydrophobic amino acids. This common feature makes it difficult for signal peptide and transmembrane helix predictors to correctly assign identity to stretches of hydrophobic residues near the N-terminal methionine of a protein sequence. The inability to reliably distinguish between N-terminal transmembrane helix and signal peptide is an error with serious consequences for the prediction of protein secretory status or transmembrane topology. In this study, we report a new method for differentiating protein N-terminal signal peptides and transmembrane helices. Based on the sequence features extracted from hydrophobic regions (amino acid frequency, hydrophobicity, and the start position), we set up discriminant functions and examined them on non-redundant datasets with jackknife tests. This method can incorporate other signal peptide prediction methods and achieve higher prediction accuracy. For Gram-negative bacterial proteins, 95.7% of N-terminal signal peptides and transmembrane helices can be correctly predicted (coefficient 0.90). Given a sensitivity of 90%, transmembrane helices can be identified from signal peptides with a precision of 99% (coefficient 0.92). For eukaryotic proteins, 94.2% of N-terminal signal peptides and transmembrane helices can be correctly predicted with coefficient 0.83. Given a sensitivity of 90%, transmembrane helices can be identified from signal peptides with a precision of 87% (coefficient 0.85). The method can be used to complement current transmembrane protein prediction and signal peptide prediction methods to improve their prediction accuracies.     

Comprehensive analysis of the numbers,lengths and amino acid compositions of transmembrane helices in prokaryotic,eukaryotic and viral integral membrane proteins of high-resolution structure

We report a comprehensive analysis of the numbers, lengths and amino acid compositions of transmembrane helices in 235 high-resolution structures of integral membrane proteins. The properties of 1551 transmembrane helices in the structures were compared with those obtained by analysis of the same amino acid sequences using topology prediction tools. Explanations for the 81 (5.2%) missing or additional transmembrane helices in the prediction results were identified. Main reasons for missing transmembrane helices were mis-identification of N-terminal signal peptides, breaks in α-helix conformation or charged residues in the middle of transmembrane helices and transmembrane helices with unusual amino acid composition. The main reason for additional transmembrane helices was mis-identification of amphipathic helices, extramembrane helices or hairpin re-entrant loops. Transmembrane helix length had an overall median of 24 residues and an average of 24.9 ± 7.0 residues and the most common length was 23 residues. The overall content of residues in transmembrane helices as a percentage of the full proteins had a median of 56.8% and an average of 55.7 ± 16.0%. Amino acid composition was analysed for the full proteins, transmembrane helices and extramembrane regions. Individual proteins or types of proteins with transmembrane helices containing extremes in contents of individual amino acids or combinations of amino acids with similar physicochemical properties were identified and linked to structure and/or function. In addition to overall median and average values, all results were analysed for proteins originating from different types of organism (prokaryotic, eukaryotic, viral) and for subgroups of receptors, channels, transporters and others.     

Combining evolutionary information and neural networks to predict protein secondary structure

Using evolutionary information contained in multiple sequence alignments as input to neural networks, secondary structure can be predicted at significantly increased accuracy. Here, we extend our previous three-level system of neural networks by using additional input information derived from multiple alignments. Using a position-specific conservation weight as part of the input increases performance. Using the number of insertions and deletions reduces the tendency for overprediction and increases overall accuracy. Addition of the global amino acid content yields a further improvement, mainly in predicting structural class. The final network system has a sustained overall accuracy of 71.6% in a multiple cross-validation test on 126 unique protein chains. A test on a new set of 124 recently solved protein structures that have no significant sequence similarity to the learning set confirms the high level of accuracy. The average cross-validated accuracy for all 250 sequence-unique chains is above 72%. Using various data sets, the method is compared to alternative prediction methods, some of which also use multiple alignments: the performance advantage of the network system is at least 6 percentage points in three-state accuracy. In addition, the network estimates secondary structure content from multiple sequence alignments about as well as circular dichroism spectroscopy on a single protein and classifies 75% of the 250 proteins correctly into one of four protein structural classes. Of particular practical importance is the definition of a position-specific reliability index. For 40% of all residues the method has a sustained three-state accuracy of 88%, as high as the overall average for homology modelling. A further strength of the method is greatly increased accuracy in predicting the placement of secondary structure segments. © 1994 Wiley-Liss, Inc.     

Characteristics affecting expression and solubilization of yeast membrane proteins

Biochemical and structural analysis of membrane proteins often critically depends on the ability to overexpress and solubilize them. To identify properties of eukaryotic membrane proteins that may be predictive of successful overexpression, we analyzed expression levels of the genomic complement of over 1000 predicted membrane proteins in a recently completed Saccharomyces cerevisiae protein expression library. We detected statistically significant positive and negative correlations between high membrane protein expression and protein properties such as size, overall hydrophobicity, number of transmembrane helices, and amino acid composition of transmembrane segments. Although expression levels of membrane and soluble proteins exhibited similar negative correlations with overall hydrophobicity, high-level membrane protein expression was positively correlated with the hydrophobicity of predicted transmembrane segments. To further characterize yeast membrane proteins as potential targets for structure determination, we tested the solubility of 122 of the highest expressed yeast membrane proteins in six commonly used detergents. Almost all the proteins tested could be solubilized using a small number of detergents. Solubility in some detergents depended on protein size, number of transmembrane segments, and hydrophobicity of predicted transmembrane segments. These results suggest that bioinformatic approaches may be capable of identifying membrane proteins that are most amenable to overexpression and detergent solubilization for structural and biochemical analyses. Bioinformatic approaches could also be used in the redesign of proteins that are not intrinsically well-adapted to such studies.     

Combining hydrophobicity and helicity: a novel approach to membrane protein structure prediction

In spite of the overwhelming numbers and critical biological functions of membrane proteins, only a few have been characterized by high-resolution structural techniques. From the structures that are known, it is seen that their transmembrane (TM) segments tend to fold most often into alpha-helices. To evaluate systematically the features of these TM segments, we have taken two approaches: (1) using the experimentally-measured residence behavior of specifically designed hydrophobic peptides in RP-HPLC, a scale was derived based directly on the properties of individual amino acids incorporated into membrane-interactive helices: and (2) the relative alpha-helical propensity of each of the 20 amino acids was measured in the organic non-polar environment of n-butanol. By combining the resulting hydrophobicity and helical propensity data, in conjunction with consideration of the 'threshold hydrophobicity' required for spontaneous membrane integration of protein segments, an approach was developed for prediction of TM segments wherein each must fulfill the dual requirements of hydrophobicity and helicity. Evaluated against the available high-resolution structural data on membrane proteins, the present combining method is shown to provide accurate predictions for the locations of TM helices. In contrast, no segment in soluble proteins was predicted as a 'TM helix'.     

A Continuum theory for the prediction of lateral and rotational positioning of α-helices in membrane proteins: Bacteriorhodopsin

We have developed a new method for the prediction of the lateral and the rotational positioning of transmembrane helices, based upon the present status of knowledge about the dominant interaction of the tertiary structure formation. The basic assumption about the interaction is that the interhelix binding is due to the polar interactions and that very short extramembrane loop segments restrict the relative position of the helices. Another assumption is made for the simplification of the prediction that a helix may be regarded as a continuum rod having polar interaction fields around it. The polar interaction field is calculated by a probe helix method, using a copolymer of serine and alanine as probe helices. The lateral position of helices is determined by the strength of the interhelix binding estimated from the polar interaction field together with the length of linking loop segments. The rotational positioning is determined by the polar interaction field, assuming the optimum lateral configuration. The structural change due to the binding of a prosthetic group is calculated, fixing the rotational freedom of a helix that is connected to the prosthetic group. Applying this method to bacteriorhodopsin, the optimum lateral and rotational positioning of transmembrane helices that are very similar to the experimental configuration was obtained. This method was implemented by a software system, which was developed for this work, and automatic calculation became possible for membrane proteins comprised of several transmembrane helices. © 1995 Wiley-Liss, Inc.     

Strand-loop-strand motifs: prediction of hairpins and diverging turns in proteins

Beta-sheet proteins have been particularly challenging for de novo structure prediction methods, which tend to pair adjacent beta-strands into beta-hairpins and produce overly local topologies. To remedy this problem and facilitate de novo prediction of beta-sheet protein structures, we have developed a neural network that classifies strand-loop-strand motifs by local hairpins and nonlocal diverging turns by using the amino acid sequence as input. The neural network is trained with a representative subset of the Protein Data Bank and achieves a prediction accuracy of 75.9 +/- 4.4% compared to a baseline prediction rate of 59.1%. Hairpins are predicted with an accuracy of 77.3 +/- 6.1%, diverging turns with an accuracy of 73.9 +/- 6.0%. Incorporation of the beta-hairpin/diverging turn classification into the ROSETTA de novo structure prediction method led to higher contact order models and somewhat improved tertiary structure predictions for a test set of 11 all-beta-proteins and 3 alphabeta-proteins. The beta-hairpin/diverging turn classification from amino acid sequences is available online for academic use (Meiler and Kuhn, 2003; www.jens-meiler.de/turnpred.html).     

Hidden markov model for the prediction of transmembrane proteins using MATLAB

Since membranous proteins play a key role in drug targeting therefore transmembrane proteins prediction is active and challenging area of biological sciences. Location based prediction of transmembrane proteins are significant for functional annotation of protein sequences. Hidden markov model based method was widely applied for transmembrane topology prediction. Here we have presented a revised and a better understanding model than an existing one for transmembrane protein prediction. Scripting on MATLAB was built and compiled for parameter estimation of model and applied this model on amino acid sequence to know the transmembrane and its adjacent locations. Estimated model of transmembrane topology was based on TMHMM model architecture. Only 7 super states are defined in the given dataset, which were converted to 96 states on the basis of their length in sequence. Accuracy of the prediction of model was observed about 74 %, is a good enough in the area of transmembrane topology prediction. Therefore we have concluded the hidden markov model plays crucial role in transmembrane helices prediction on MATLAB platform and it could also be useful for drug discovery strategy. AVAILABILITY: The database is available for free at bioinfonavneet@gmail.comvinaysingh@bhu.ac.in.     

Genetic mapping of the human C5a receptor. Identification of transmembrane amino acids critical for receptor function

Many hormones and sensory stimuli signal through a superfamily of seven transmembrane-spanning receptors to activate heterotrimeric G proteins. How the seven transmembrane segments of the receptors (a molecular architecture of bundled alpha-helices conserved from yeast to man) work as "on/off" switches remains unknown. Previously, we used random saturation mutagenesis coupled with a genetic selection in yeast to determine the relative importance of amino acids in four of the seven transmembrane segments of the human C5a receptor (Baranski, T. J., Herzmark, P., Lichtarge, O., Gerber, B. O., Trueheart, J., Meng, E. C., Iiri, T., Sheikh, S. P., and Bourne, H. R. (1999) J. Biol. Chem. 274, 15757-15765). In this study, we evaluate helices I, II, and IV, thereby furnishing a complete mutational map of the seven transmembrane helices of the human C5a receptor. Our analysis identified 19 amino acid positions resistant to non-conservative substitutions. When combined with the 25 essential residues previously identified in helices III and V-VII, they delineate two distinct components of the receptor switch: a ligand-binding surface at or near the extracellular surface of the helix bundle and a core cluster in the cytoplasmic half of the bundle. In addition, we found critical amino acids in the first and second helices that are predicted to face the lipid membrane. These residues form an extended surface that might mediate interactions with lipids and other membrane proteins or function as an oligomerization domain with other receptors.     

Discriminating lysosomal membrane protein types using dynamic neural network

This work presents a dynamic artificial neural network methodology, which classifies the proteins into their classes from their sequences alone: the lysosomal membrane protein classes and the various other membranes protein classes. In this paper, neural networks-based lysosomal-associated membrane protein type prediction system is proposed. Different protein sequence representations are fused to extract the features of a protein sequence, which includes seven feature sets; amino acid (AA) composition, sequence length, hydrophobic group, electronic group, sum of hydrophobicity, R-group, and dipeptide composition. To reduce the dimensionality of the large feature vector, we applied the principal component analysis. The probabilistic neural network, generalized regression neural network, and Elman regression neural network (RNN) are used as classifiers and compared with layer recurrent network (LRN), a dynamic network. The dynamic networks have memory, i.e. its output depends not only on the input but the previous outputs also. Thus, the accuracy of LRN classifier among all other artificial neural networks comes out to be the highest. The overall accuracy of jackknife cross-validation is 93.2% for the data-set. These predicted results suggest that the method can be effectively applied to discriminate lysosomal associated membrane proteins from other membrane proteins (Type-I, Outer membrane proteins, GPI-Anchored) and Globular proteins, and it also indicates that the protein sequence representation can better reflect the core feature of membrane proteins than the classical AA composition.     

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.     

氨基酸邻域对二级结构的影响程度研究

通过研究神经网络权值矩阵的算法,挖掘蛋白质二级结构与氨基酸序列间的内在规律,提高一级序列预测二级结构的准确度。神经网络方法在特征分类方面具有良好表现,经过学习训练后的神经元连接权值矩阵包含样本的内在特征和规律。研究使用神经网络权值矩阵打分预测;采用错位比对方法寻找敏感的氨基酸邻域;分析测试集在不同加窗长度下的共性表现。实验表明,在滑动窗口长度L=7时,预测性能变化显著;邻域位置P=4的氨基酸残基对预测性能有加强作用。该研究方法为基于局部序列特征的蛋白质二级结构预测提供了新的算法设计。     

Conformational preference functions for predicting helices in membrane proteins

A suite of FORTRAN programs, PREF, is described for calculating preference functions from the data base of known protein structures and for comparing smoothed profiles of sequence-dependent preferences in proteins of unknown structure. Amino acid preferences for a secondary structure are considered as functions of a sequence environment. Sequence environment of amino acid residue in a protein is defined as an average over some physical, chemical, or statistical property of its primary structure neighbors. The frequency distribution of sequence environments in the data base of soluble protein structures is approximately normal for each amino acid type of known secondary conformation. An analytical expression for the dependence of preferences on sequence environment is obtained after each frequency distribution is replaced by corresponding Gaussian function. The preference for the α-helical conformation increases for each amino acid type with the increase of sequence environment of buried solvent-accessible surface areas. We show that a set of preference functions based on buried surface area is useful for predicting folding motifs in α-class proteins and in integral membrane proteins. The prediction accuracy for helical residues is 79% for 5 integral membrane proteins and 74% for 11 α-class soluble proteins. Most residues found in transmembrane segments of membrane proteins with known α-helical structure are predicted to be indeed in the helical conformation because of very high middle helix preferences. Both extramembrane and transmembrane helices in the photosynthetic reaction center M and L subunits are correctly predicted. We point out in the discussion that our method of conformational preference functions can identify what physical properties of the amino acids are important in the formation of particular secondary structure elements. © 1993 John Wiley & Sons, Inc.     

Topology of a human equilibrative, nitrobenzylthioinosine (NBMPR)-sensitive nucleoside transporter (hENT1) implicated in the cellular uptake of adenosine and anti-cancer drugs

The human equilibrative nucleoside transporter hENT1, the first identified member of the ENT family of integral membrane proteins, is the primary mechanism for the cellular uptake of physiologic nucleosides, including adenosine, and many anti-cancer nucleoside drugs. We have produced recombinant hENT1 in Xenopus oocytes and used native and engineered N-glycosylation sites in combination with immunological approaches to experimentally define the membrane architecture of this prototypic nucleoside transporter. hENT1 (456 amino acid residues) is shown to contain 11 transmembrane helical segments with an amino terminus that is intracellular and a carboxyl terminus that is extracellular. Transmembrane helices are linked by short hydrophilic regions, except for a large glycosylated extracellular loop between transmembrane helices 1 and 2 and a large central cytoplasmic loop between transmembrane helices 6 and 7. Sequence analyses suggest that this membrane topology is common to all mammalian, insect, nematode, protozoan, yeast, and plant members of the ENT protein family.     

Predicting the helix packing of globular proteins by self-correcting distance geometry.

A new self-correcting distance geometry method for predicting the three-dimensional structure of small globular proteins was assessed with a test set of 8 helical proteins. With the knowledge of the amino acid sequence and the helical segments, our completely automated method calculated the correct backbone topology of six proteins. The accuracy of the predicted structures ranged from 2.3 A to 3.1 A for the helical segments compared to the experimentally determined structures. For two proteins, the predicted constraints were not restrictive enough to yield a conclusive prediction. The method can be applied to all small globular proteins, provided the secondary structure is known from NMR analysis or can be predicted with high reliability.