Bio-support vector machines for computational proteomics

MOTIVATION: One of the most important issues in computational proteomics is to produce a prediction model for the classification or annotation of biological function of novel protein sequences. In order to improve the prediction accuracy, much attention has been paid to the improvement of the performance of the algorithms used, few is for solving the fundamental issue, namely, amino acid encoding as most existing pattern recognition algorithms are unable to recognize amino acids in protein sequences. Importantly, the most commonly used amino acid encoding method has the flaw that leads to large computational cost and recognition bias. RESULTS: By replacing kernel functions of support vector machines (SVMs) with amino acid similarity measurement matrices, we have modified SVMs, a new type of pattern recognition algorithm for analysing protein sequences, particularly for proteolytic cleavage site prediction. We refer to the modified SVMs as bio-support vector machine. When applied to the prediction of HIV protease cleavage sites, the new method has shown a remarkable advantage in reducing the model complexity and enhancing the model robustness.     

Neural network-based prediction of mutation-induced protein stability changes in Staphylococcal nuclease at 20 residue positions

Protein-based therapeutics are playing an increasingly important role in the treatment of diseases, including diabetes and cancer. The viability of these treatments, however, are highly dependent on the stability of the therapeutic, since stability affects both the shelf life of the therapeutic as well as its active life in the body. Stability engineering can, therefore, be used to increase the effectiveness of protein-based therapeutics. Computational methods of protein stability prediction have been under development for about a decade, but complex molecular interactions make stability prediction difficult and computationally intensive. A rapid computational method of protein stability prediction is developed using feed-forward neural networks and used to predict mutation-induced stability changes in Staphylococcal nuclease. The input to the neural network consisted of sequences of evolutionarily based amino acid similarity scores that were obtained through the comparison of the amino acids in a mutation containing sequence to their positional counterparts in the baseline wild-type amino acid sequence. A training set was created which consisted of similarity score sequences, for which the stabilities of the corresponding amino acid sequences were known, paired with the relative stabilities of the sequences to that of the baseline. Back-propagation of error was used to train the network to output accurate relative stability scores for the sequences in the training set. Neural network-based relative stability predictions for 55 sequences containing mutation combinations not found in the training set had an accuracy of 92.8%.     

Predicting disordered regions in proteins based on decision trees of reduced amino acid composition.

Intrinsically unstructured proteins (IUPs) are proteins lacking a fixed three dimensional structure or containing long disordered regions. IUPs play an important role in biology and disease. Identifying disordered regions in protein sequences can provide useful information on protein structure and function, and can assist high-throughput protein structure determination. In this paper we present a system for predicting disordered regions in proteins based on decision trees and reduced amino acid composition. Concise rules based on biochemical properties of amino acid side chains are generated for prediction. Coarser information extracted from the composition of amino acids can not only improve the prediction accuracy but also increase the learning efficiency. In cross-validation tests, with four groups of reduced amino acid composition, our system can achieve a recall of 80% at a 13% false positive rate for predicting disordered regions, and the overall accuracy can reach 83.4%. This prediction accuracy is comparable to most, and better than some, existing predictors. Advantages of our approach are high prediction accuracy for long disordered regions and efficiency for large-scale sequence analysis. Our software is freely available for academic use upon request.     

Prediction of protein folding class from amino acid composition

An empirical relation between the amino acid composition and three-dimensional folding pattern of several classes of proteins has been determined. Computer simulated neural networks have been used to assign proteins to one of the following classes based on their amino acid composition and size: (1) 4α-helical bundles, (2) parallel (α/β)₈ barrels, (3) nucleotide binding fold, (4) immunoglobulin fold, or (5) none of these. Networks trained on the known crystal structures as well as sequences of closely related proteins are shown to correctly predict folding classes of proteins not represented in the training set with an average accuracy of 87%. Other folding motifs can easily be added to the prediction scheme once larger databases become available. Analysis of the neural network weights reveals that amino acids favoring prediction of a folding class are usually over represented in that class and amino acids with unfavorable weights are underrepresented in composition. The neural networks utilize combinations of these multiple small variations in amino acid composition in order to make a prediction. The favorably weighted amino acids in a given class also form the most intramolecular interactions with other residues in proteins of that class. A detailed examination of the contacts of these amino acids reveals some general patterns that may help stabilize each folding class. © 1993 Wiley-Liss, Inc.     

Prediction of pattern recognition receptor family using pseudo-amino acid composition

Pattern recognition receptors (PRRs) play a key role in the innate immune response by recognizing pathogen associated molecular patterns derived from a diverse collection of microbial pathogens. PRRs form a superfamily of proteins related to host health and disease. Thus, prediction of PRR family might supply biologically significant information for functional annotation of PRRs and development of novel drugs. In this paper, a computational method is proposed for predicting the families of PRRs. The prediction was performed on the basis of amino acid composition and pseudo-amino acid composition (PseAAC) from primary sequences of proteins using support vector machines. A non-redundant dataset consisted of 332 PRRs in seven families was constructed to do training and testing. It was demonstrated that different families of PRRs were quite closely correlated with amino acid composition as well as PseAAC. In the jackknife test, overall accuracies of amino acid composition-based and PseAAC-based classifiers reached 96.1% and 97.9%, respectively. The results indicate that families of PRRs are predictable with high accuracy. It is anticipated that this computational method might be a powerful tool for the automated assignment of families of PRRs.     

Accuracy and power of bayes prediction of amino acid sites under positive selection

Bayes prediction quantifies uncertainty by assigning posterior probabilities. It was used to identify amino acids in a protein under recurrent diversifying selection indicated by higher nonsynonymous (d(N)) than synonymous (d(S)) substitution rates or by omega = d(N)/d(S) > 1. Parameters were estimated by maximum likelihood under a codon substitution model that assumed several classes of sites with different omega ratios. The Bayes theorem was used to calculate the posterior probabilities of each site falling into these site classes. Here, we evaluate the performance of Bayes prediction of amino acids under positive selection by computer simulation. We measured the accuracy by the proportion of predicted sites that were truly under selection and the power by the proportion of true positively selected sites that were predicted by the method. The accuracy was slightly better for longer sequences, whereas the power was largely unaffected by the increase in sequence length. Both accuracy and power were higher for medium or highly diverged sequences than for similar sequences. We found that accuracy and power were unacceptably low when data contained only a few highly similar sequences. However, sampling a large number of lineages improved the performance substantially. Even for very similar sequences, accuracy and power can be high if over 100 taxa are used in the analysis. We make the following recommendations: (1) prediction of positive selection sites is not feasible for a few closely related sequences; (2) using a large number of lineages is the best way to improve the accuracy and power of the prediction; and (3) multiple models of heterogeneous selective pressures among sites should be applied in real data analysis.     

Kinking the coiled coil--negatively charged residues at the coiled-coil interface

The coiled coil is one of the most common protein-structure motifs. It is believed to be adopted by 3-5% of all amino acids in proteins. It comprises two or more alpha-helical chains wrapped around one another. The sequences of most coiled coils are characterized by a seven-residue (heptad) repeat, denoted (abcdefg)(n). Residues at the a and d positions define the helical interface (core) and are usually hydrophobic, though about 20% are polar or charged. We show that parallel coiled-coils have a unique pattern of their negatively charged residues at the core positions: aspartic acid is excluded from these positions while glutamic acid is not. In contrast the antiparallel structures are more permissive in their amino acid usage. We show further, and for the first time, that incorporation of Asp but not Glu into the a positions of a parallel coiled coil creates a flexible hinge and that the maximal hinge angle is being directly related to the number of incorporated mutations. These new computational and experimental observations will be of use in improving protein-structure predictions, and as rules to guide rational design of novel coiled-coil motifs and coiled coil-based materials.     

Prediction of protein subcellular locations using a new measure of information discrepancy

Given a raw protein sequence, knowing its subcellular location is an important step toward understanding its function and designing further experiments. A novel method is proposed for the prediction of protein subcellular locations from sequences. For four categories of eukaryotic proteins the overall predictive accuracy is 82.0%, 2.6% higher than that by using SVM approach. For three subcellular locations of prokaryotic proteins, an overall accuracy of 89.9% is obtained. In accordance with the architecture of cells, a hierarchical prediction approach is designed. Based on amino acid composition extracellular proteins and intracellular proteins can be identified with accuracy of 97%.     

Sequence fingerprints distinguish erroneous from correct predictions of intrinsically disordered protein regions

More than 60 prediction methods for intrinsically disordered proteins (IDPs) have been developed over the years, many of which are accessible on the World Wide Web. Nearly, all of these predictors give balanced accuracies in the ~65%–~80% range. Since predictors are not perfect, further studies are required to uncover the role of amino acid residues in native IDP as compared to predicted IDP regions. In the present work, we make use of sequences of 100% predicted IDP regions, false positive disorder predictions, and experimentally determined IDP regions to distinguish the characteristics of native versus predicted IDP regions. A higher occurrence of asparagine is observed in sequences of native IDP regions but not in sequences of false positive predictions of IDP regions. The occurrences of certain combinations of amino acids at the pentapeptide level provide a distinguishing feature in the IDPs with respect to globular proteins. The distinguishing features presented in this paper provide insights into the sequence fingerprints of amino acid residues in experimentally determined as compared to predicted IDP regions. These observations and additional work along these lines should enable the development of improvements in the accuracy of disorder prediction algorithm.     

Development of METAL‐ACTIVE SITE and ZINCCLUSTER tool to predict active site pockets

The advent of whole genome sequencing leads to increasing number of proteins with known amino acid sequences. Despite many efforts, the number of proteins with resolved three dimensional structures is still low. One of the challenging tasks the structural biologists face is the prediction of the interaction of metal ion with any protein for which the structure is unknown. Based on the information available in Protein Data Bank, a site (METALACTIVE INTERACTION) has been generated which displays information for significant high preferential and low‐preferential combination of endogenous ligands for 49 metal ions. User can also gain information about the residues present in the first and second coordination sphere as it plays a major role in maintaining the structure and function of metalloproteins in biological system. In this paper, a novel computational tool (ZINCCLUSTER) is developed, which can predict the zinc metal binding sites of proteins even if only the primary sequence is known. The purpose of this tool is to predict the active site cluster of an uncharacterized protein based on its primary sequence or a 3D structure. The tool can predict amino acids interacting with a metal or vice versa. This tool is based on the occurrence of significant triplets and it is tested to have higher prediction accuracy when compared to that of other available techniques.     

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。     

Bi-Factor Analysis Based on Noise-Reduction (BIFANR): A New Algorithm for Detecting Coevolving Amino Acid Sites in Proteins

Previous statistical analyses have shown that amino acid sites in a protein evolve in a correlated way instead of independently. Even though located distantly in the linear sequence, the coevolved amino acids could be spatially adjacent in the tertiary structure, and constitute specific protein sectors. Moreover, these protein sectors are independent of one another in structure, function, and even evolution. Thus, systematic studies on protein sectors inside a protein will contribute to the clarification of protein function. In this paper, we propose a new algorithm BIFANR (Bi-factor Analysis Based on Noise-reduction) for detecting protein sectors in amino acid sequences. After applying BIFANR on S1A family and PDZ family, we carried out internal correlation test, statistical independence test, evolutionary rate analysis, evolutionary independence analysis, and function analysis to assess the prediction. The results showed that the amino acids in certain predicted protein sector are closely correlated in structure, function, and evolution, while protein sectors are nearly statistically independent. The results also indicated that the protein sectors have distinct evolutionary directions. In addition, compared with other algorithms, BIFANR has higher accuracy and robustness under the influence of noise sites.     

基于位置特异性谱和输入加权神经网络的蛋白质亚细胞定位预测

蛋白质必须处于正确的亚细胞位置才能行使其功能。文章利用PSI-BLAST工具搜索蛋白质序列,提取位点特异性谱中的位点特异性得分矩阵作为蛋白质的一类特征,并计算4等分序列的氨基酸含量以及1～7阶二肽含量作为另外两类特征,由这三类特征一共得到蛋白质序列的12个特征向量。通过设计一个简单加权函数对各类特征向量加权处理,作为神经网络预测器的输入,并使用Levenberg-Marquardt算法代替传统的EBP算法来调整网络权值和阈值,大大提高了训练速度。对具有4类亚细胞位置和12类亚细胞位置的两种蛋白质数据集分别进行＂留一法＂测试和5倍交叉验证测试,总体预测精度分别达到88.4%和83.3%。其中,对4类亚细胞位置数据集的预测效果优于普通BP神经网络、隐马尔可夫模型、模糊K邻近等预测方法,对12类亚细胞位置数据集的预测效果优于支持向量机分类方法。最后还对三类特征采取不同加权比例对预测精度的影响进行了讨论,对选择的八种加权比例的预测结果表明,分别给予三类特征合适的权值系数可以进一步提高预测精度。     

Prediction of protein secondary structure content using amino acid composition and evolutionary information

Knowing protein structure and inferring its function from the structure are one of the main issues of computational structural biology, and often the first step is studying protein secondary structure. There have been many attempts to predict protein secondary structure contents. Previous attempts assumed that the content of protein secondary structure can be predicted successfully using the information on the amino acid composition of a protein. Recent methods achieved remarkable prediction accuracy by using the expanded composition information. The overall average error of the most successful method is 3.4%. Here, we demonstrate that even if we only use the simple amino acid composition information alone, it is possible to improve the prediction accuracy significantly if the evolutionary information is included. The idea is motivated by the observation that evolutionarily related proteins share the similar structure. After calculating the homolog-averaged amino acid composition of a protein, which can be easily obtained from the multiple sequence alignment by running PSI-BLAST, those 20 numbers are learned by a multiple linear regression, an artificial neural network and a support vector regression. The overall average error of method by a support vector regression is 3.3%. It is remarkable that we obtain the comparable accuracy without utilizing the expanded composition information such as pair-coupled amino acid composition. This work again demonstrates that the amino acid composition is a fundamental characteristic of a protein. It is anticipated that our novel idea can be applied to many areas of protein bioinformatics where the amino acid composition information is utilized, such as subcellular localization prediction, enzyme subclass prediction, domain boundary prediction, signal sequence prediction, and prediction of unfolded segment in a protein sequence, to name a few.     

Critical amino acids responsible for conferring calcium channel characteristics are located on the surface and aroundβ-turn potentials of channel proteins

Calcium ion is thought to be one of the initial signals in the process of synaptic modification. Various reports have described that the critical amino acids responsible for determining calcium permeability of ion channels are glutamic acid, glutamine, arginine, and asparagine. By using a computational method (MacPROT) distinguishing transmembrane, globular, and surface sequences of proteins, the present work predicts that the critical amino acids exist within surface regions of the proteins. Furthermore, occurrence ofβ-turn probabilities can be predicted around these critical residues by the protein conformational prediction method of Chou and Fasman. The results suggest that the critical amino acids exist at hydrophilic spaces or canals of membranous channel proteins and that the redirection potential of the protein chain induced by the turn structures provides the conformational change requisite for the ion selectivity and gating (opening/closing) of the channels.     

Predicting the protein SUMO modification sites based on Properties Sequential Forward Selection (PSFS)

Protein SUMO modification is an important post-translational modification and the optimization of prediction methods remains a challenge. Here, by using Support Vector Machines algorithm (SVM), a novel computational method was developed for SUMO modification site prediction based on Sequential Forward Selection (SFS) of hundreds of amino acid properties, which are collected by Amino Acid Index database (http://www.genome.jp/aaindex). Our method also compares with the 0/1 system, in which the 20 amino acids are represented by 20-dimensional vectors (A = 00000000000000000001, C = 00000000000000000010 and so on). The overall accuracy of leave-one-out cross-validation for our method reaches 89.18%, which is higher than 0/1 system. It indicated that the SUMO modification prediction process is highly related to the amino acid property and this approach here provide a helpful tool for further investigation of the SUMO modification and identification of sumoylation sites in proteins. The software is available at http://www.biosino.org/sumo.     

Prediction of protein surface accessibility with information theory

A new, simple method based on information theory is introduced to predict the solvent accessibility of amino acid residues in various states defined by their different thresholds. Prediction is achieved by the application of information obtained from a single amino acid position or pair-information for a window of seventeen amino acids around the desired residue. Results obtained by pairwise information values are better than results from single amino acids. This reinforces the effect of the local environment on the accessibility of amino acid residues. The prediction accuracy of this method in a jackknife test system for two and three states is better than 70 and 60 %, respectively. A comparison of the results with those reported by others involving the same data set also testifies to a better prediction accuracy in our case.     

Prediction of phosphorylation sites using SVMs

MOTIVATION: Phosphorylation is involved in diverse signal transduction pathways. By predicting phosphorylation sites and their kinases from primary protein sequences, we can obtain much valuable information that can form the basis for further research. Using support vector machines, we attempted to predict phosphorylation sites and the type of kinase that acts at each site. RESULTS: Our prediction system was limited to phosphorylation sites catalyzed by four protein kinase families and four protein kinase groups. The accuracy of the predictions ranged from 83 to 95% at the kinase family level, and 76-91% at the kinase group level. The prediction system used-PredPhospho-can be applied to the functional study of proteins, and can help predict the changes in phosphorylation sites caused by amino acid variations at intra- and interspecies levels.     

用离散量的方法识别蛋白质的超二级结构

用离散量的方法,对2208个分辨率在2.5I以上的高精度的蛋白质结构中四类超二级结构进行了识别。从蛋白质一级序列出发,以氨基酸(20种氨基酸加一个空位)和其紧邻关联共同为参数,当序列模式固定长取8个氨基酸残基时,对“822”序列模式3交叉检验的平均预测精度达到78.1%,jack-knife检验的平均预测精度达到76.7%;当序列模式固定长取10个氨基酸残基时,对“1041”序列模式3交叉检验的平均预测精度达到83.1%,jack-knife检验的平均预测精度达到79.8%。