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.     

Prediction of turn types in protein structure by machine-learning classifiers

We present machine learning approaches for turn prediction from the amino acid sequence. Different turn classes and types were considered based on a novel turn classification scheme. We trained an unsupervised (self-organizing map) and two kernel-based classifiers, namely the support vector machine and a probabilistic neural network. Turn versus non-turn classification was carried out for turn families containing intramolecular hydrogen bonds and three to six residues. Support vector machine classifiers yielded a Matthews correlation coefficient (mcc) of approximately 0.6 and a prediction accuracy of 80%. Probabilistic neural networks were developed for beta-turn type prediction. The method was able to distinguish between five types of beta-turns yielding mcc > 0.5 and at least 80% overall accuracy. We conclude that the proposed new turn classification is distinct and well-defined, and machine learning classifiers are suited for sequence-based turn prediction. Their potential for sequence-based prediction of turn structures is discussed.     

Mem-PHybrid: hybrid features-based prediction system for classifying membrane protein types

Membrane proteins are a major class of proteins and encoded by approximately 20% to 30% of genes in most organisms. In this work, a two-layer novel membrane protein prediction system, called Mem-PHybrid, is proposed. It is able to first identify the protein query as a membrane or nonmembrane protein. In the second level, it further identifies the type of membrane protein. The proposed Mem-PHybrid prediction system is based on hybrid features, whereby a fusion of both the physicochemical and split amino acid composition-based features is performed. This enables the proposed Mem-PHybrid to exploit the discrimination capabilities of both types of feature extraction strategy. In addition, minimum redundancy and maximum relevance has also been applied to reduce the dimensionality of a feature vector. We employ random forest, evidence-theoretic K-nearest neighbor, and support vector machine (SVM) as classifiers and analyze their performance on two datasets. SVM using hybrid features yields the highest accuracy of 89.6% and 97.3% on dataset1 and 91.5% and 95.5% on dataset2 for jackknife and independent dataset tests, respectively. The enhanced prediction performance of Mem-PHybrid is largely attributed to the exploitation of the discrimination power of the hybrid features and of the learning capability of SVM. Mem-PHybrid is accessible at http://www.111.68.99.218/Mem-PHybrid.     

Predicting protein subcellular location: exploiting amino acid based sequence of feature spaces and fusion of diverse classifiers

A novel approach CE-Ploc is proposed for predicting protein subcellular locations by exploiting diversity both in feature and decision spaces. The diversity in a sequence of feature spaces is exploited using hydrophobicity and hydrophilicity of amphiphilic pseudo amino acid composition and a specific learning mechanism. Diversity in learning mechanisms is exploited by fusion of classifiers that are based on different learning mechanisms. Significant improvement in prediction performance is observed using jackknife and independent dataset tests.     

A deep dense inception network for protein beta-turn prediction

Beta-turn prediction is useful in protein function studies and experimental design. Although recent approaches using machine-learning techniques such as support vector machine (SVM), neural networks, and K nearest neighbor have achieved good results for beta-turn prediction, there is still significant room for improvement. As previous predictors utilized features in a sliding window of 4-20 residues to capture interactions among sequentially neighboring residues, such feature engineering may result in incomplete or biased features and neglect interactions among long-range residues. Deep neural networks provide a new opportunity to address these issues. Here, we proposed a deep dense inception network (DeepDIN) for beta-turn prediction, which takes advantage of the state-of-the-art deep neural network design of dense networks and inception networks. A test on a recent BT6376 benchmark data set shows that DeepDIN outperformed the previous best tool BetaTPred3 significantly in both the overall prediction accuracy and the nine-type beta-turn classification accuracy. A tool, called MUFold-BetaTurn, was developed, which is the first beta-turn prediction tool utilizing deep neural networks. The tool can be downloaded at http://dslsrv8.cs.missouri.edu/~cf797/MUFoldBetaTurn/download.html .     

A novel protein structural classes prediction method based on predicted secondary structure

Knowledge of structural classes plays an important role in understanding protein folding patterns. In this paper, features based on the predicted secondary structure sequence and the corresponding E–H sequence are extracted. Then, an 11-dimensional feature vector is selected based on a wrapper feature selection algorithm and a support vector machine (SVM). Among the 11 selected features, 4 novel features are newly designed to model the differences between α/β class and α + β class, and other 7 rational features are proposed by previous researchers. To examine the performance of our method, a total of 5 datasets are used to design and test the proposed method. The results show that competitive prediction accuracies can be achieved by the proposed method compared to existing methods (SCPRED, RKS-PPSC and MODAS), and 4 new features are demonstrated essential to differentiate α/β and α + β classes. Standalone version of the proposed method is written in JAVA language and it can be downloaded from http://web.xidian.edu.cn/slzhang/paper.html.     

A novel method of protein secondary structure prediction with high segment overlap measure: support vector machine approach

We have introduced a new method of protein secondary structure prediction which is based on the theory of support vector machine (SVM). SVM represents a new approach to supervised pattern classification which has been successfully applied to a wide range of pattern recognition problems, including object recognition, speaker identification, gene function prediction with microarray expression profile, etc. In these cases, the performance of SVM either matches or is significantly better than that of traditional machine learning approaches, including neural networks.The first use of the SVM approach to predict protein secondary structure is described here. Unlike the previous studies, we first constructed several binary classifiers, then assembled a tertiary classifier for three secondary structure states (helix, sheet and coil) based on these binary classifiers. The SVM method achieved a good performance of segment overlap accuracy SOV=76.2 % through sevenfold cross validation on a database of 513 non-homologous protein chains with multiple sequence alignments, which out-performs existing methods. Meanwhile three-state overall per-residue accuracy Q(3) achieved 73.5 %, which is at least comparable to existing single prediction methods. Furthermore a useful "reliability index" for the predictions was developed. In addition, SVM has many attractive features, including effective avoidance of overfitting, the ability to handle large feature spaces, information condensing of the given data set, etc. The SVM method is conveniently applied to many other pattern classification tasks in biology.     

Prediction of protein structural class for low-similarity sequences using support vector machine and PSI-BLAST profile

Knowledge of structural class plays an important role in understanding protein folding patterns. In this study, a simple and powerful computational method, which combines support vector machine with PSI-BLAST profile, is proposed to predict protein structural class for low-similarity sequences. The evolution information encoding in the PSI-BLAST profiles is converted into a series of fixed-length feature vectors by extracting amino acid composition and dipeptide composition from the profiles. The resulting vectors are then fed to a support vector machine classifier for the prediction of protein structural class. To evaluate the performance of the proposed method, jackknife cross-validation tests are performed on two widely used benchmark datasets, 1189 (containing 1092 proteins) and 25PDB (containing 1673 proteins) with sequence similarity lower than 40% and 25%, respectively. The overall accuracies attain 70.7% and 72.9% for 1189 and 25PDB datasets, respectively. Comparison of our results with other methods shows that our method is very promising to predict protein structural class particularly for low-similarity datasets and may at least play an important complementary role to existing methods.     

MUFOLD‐SS: New deep inception‐inside‐inception networks for protein secondary structure prediction

Protein secondary structure prediction can provide important information for protein 3D structure prediction and protein functions. Deep learning offers a new opportunity to significantly improve prediction accuracy. In this article, a new deep neural network architecture, named the Deep inception‐inside‐inception (Deep3I) network, is proposed for protein secondary structure prediction and implemented as a software tool MUFOLD‐SS. The input to MUFOLD‐SS is a carefully designed feature matrix corresponding to the primary amino acid sequence of a protein, which consists of a rich set of information derived from individual amino acid, as well as the context of the protein sequence. Specifically, the feature matrix is a composition of physio‐chemical properties of amino acids, PSI‐BLAST profile, and HHBlits profile. MUFOLD‐SS is composed of a sequence of nested inception modules and maps the input matrix to either eight states or three states of secondary structures. The architecture of MUFOLD‐SS enables effective processing of local and global interactions between amino acids in making accurate prediction. In extensive experiments on multiple datasets, MUFOLD‐SS outperformed the best existing methods and other deep neural networks significantly. MUFold‐SS can be downloaded from http://dslsrv8.cs.missouri.edu/~cf797/MUFoldSS/download.html .     

Prediction of membrane proteins using split amino acid and ensemble classification

Knowledge of the types of membrane protein provides useful clues in deducing the functions of uncharacterized membrane proteins. An automatic method for efficiently identifying uncharacterized proteins is thus highly desirable. In this work, we have developed a novel method for predicting membrane protein types by exploiting the discrimination capability of the difference in amino acid composition at the N and C terminus through split amino acid composition (SAAC). We also show that the ensemble classification can better exploit this discriminating capability of SAAC. In this study, membrane protein types are classified using three feature extraction and several classification strategies. An ensemble classifier Mem-EnsSAAC is then developed using the best feature extraction strategy. Pseudo amino acid (PseAA) composition, discrete wavelet analysis (DWT), SAAC, and a hybrid model are employed for feature extraction. The nearest neighbor, probabilistic neural network, support vector machine, random forest, and Adaboost are used as individual classifiers. The predicted results of the individual learners are combined using genetic algorithm to form an ensemble classifier, Mem-EnsSAAC yielding an accuracy of 92.4 and 92.2% for the Jackknife and independent dataset test, respectively. Performance measures such as MCC, sensitivity, specificity, F-measure, and Q-statistics show that SAAC-based prediction yields significantly higher performance compared to PseAA- and DWT-based systems, and is also the best reported so far. The proposed Mem-EnsSAAC is able to predict the membrane protein types with high accuracy and consequently, can be very helpful in drug discovery. It can be accessed at http://111.68.99.218/membrane.     

Prediction of protein binding sites in protein structures using hidden Markov support vector machine

Background  

Predicting the binding sites between two interacting proteins provides important clues to the function of a protein. Recent research on protein binding site prediction has been mainly based on widely known machine learning techniques, such as artificial neural networks, support vector machines, conditional random field, etc. However, the prediction performance is still too low to be used in practice. It is necessary to explore new algorithms, theories and features to further improve the performance.     

Cascaded multiple classifiers for secondary structure prediction

We describe a new classifier for protein secondary structure prediction that is formed by cascading together different types of classifiers using neural networks and linear discrimination. The new classifier achieves an accuracy of 76.7% (assessed by a rigorous full Jack-knife procedure) on a new nonredundant dataset of 496 nonhomologous sequences (obtained from G.J. Barton and J.A. Cuff). This database was especially designed to train and test protein secondary structure prediction methods, and it uses a more stringent definition of homologous sequence than in previous studies. We show that it is possible to design classifiers that can highly discriminate the three classes (H, E, C) with an accuracy of up to 78% for beta-strands, using only a local window and resampling techniques. This indicates that the importance of long-range interactions for the prediction of beta-strands has been probably previously overestimated.     

GPCR-MPredictor: multi-level prediction of G protein-coupled receptors using genetic ensemble

G protein-coupled receptors (GPCRs) are transmembrane proteins, which transduce signals from extracellular ligands to intracellular G protein. Automatic classification of GPCRs can provide important information for the development of novel drugs in pharmaceutical industry. In this paper, we propose an evolutionary approach, GPCR-MPredictor, which combines individual classifiers for predicting GPCRs. GPCR-MPredictor is a web predictor that can efficiently predict GPCRs at five levels. The first level determines whether a protein sequence is a GPCR or a non-GPCR. If the predicted sequence is a GPCR, then it is further classified into family, subfamily, sub-subfamily, and subtype levels. In this work, our aim is to analyze the discriminative power of different feature extraction and classification strategies in case of GPCRs prediction and then to use an evolutionary ensemble approach for enhanced prediction performance. Features are extracted using amino acid composition, pseudo amino acid composition, and dipeptide composition of protein sequences. Different classification approaches, such as k-nearest neighbor (KNN), support vector machine (SVM), probabilistic neural networks (PNN), J48, Adaboost, and Naives Bayes, have been used to classify GPCRs. The proposed hierarchical GA-based ensemble classifier exploits the prediction results of SVM, KNN, PNN, and J48 at each level. The GA-based ensemble yields an accuracy of 99.75, 92.45, 87.80, 83.57, and 96.17% at the five levels, on the first dataset. We further perform predictions on a dataset consisting of 8,000 GPCRs at the family, subfamily, and sub-subfamily level, and on two other datasets of 365 and 167 GPCRs at the second and fourth levels, respectively. In comparison with the existing methods, the results demonstrate the effectiveness of our proposed GPCR-MPredictor in classifying GPCRs families. It is accessible at .     

Machine learning techniques for protein function prediction

Proteins play important roles in living organisms, and their function is directly linked with their structure. Due to the growing gap between the number of proteins being discovered and their functional characterization (in particular as a result of experimental limitations), reliable prediction of protein function through computational means has become crucial. This paper reviews the machine learning techniques used in the literature, following their evolution from simple algorithms such as logistic regression to more advanced methods like support vector machines and modern deep neural networks. Hyperparameter optimization methods adopted to boost prediction performance are presented. In parallel, the metamorphosis in the features used by these algorithms from classical physicochemical properties and amino acid composition, up to text-derived features from biomedical literature and learned feature representations using autoencoders, together with feature selection and dimensionality reduction techniques, are also reviewed. The success stories in the application of these techniques to both general and specific protein function prediction are discussed.     

Using genetic algorithms to select most predictive protein features

Many important characteristics of proteins such as biochemical activity and subcellular localization present a challenge to machine-learning methods: it is often difficult to encode the appropriate input features at the residue level for the purpose of making a prediction for the entire protein. The problem is usually that the biophysics of the connection between a machine-learning method's input (sequence feature) and its output (observed phenomenon to be predicted) remains unknown; in other words, we may only know that a certain protein is an enzyme (output) without knowing which region may contain the active site residues (input). The goal then becomes to dissect a protein into a vast set of sequence-derived features and to correlate those features with the desired output. We introduce a framework that begins with a set of global sequence features and then vastly expands the feature space by generically encoding the coexistence of residue-based features. It is this combination of individual features, that is the step from the fractions of serine and buried (input space 20 + 2) to the fraction of buried serine (input space 20 * 2) that implicitly shifts the search space from global feature inputs to features that can capture very local evidence such as a the individual residues of a catalytic triad. The vast feature space created is explored by a genetic algorithm (GA) paired with neural networks and support vector machines. We find that the GA is critical for selecting combinations of features that are neither too general resulting in poor performance, nor too specific, leading to overtraining. The final framework manages to effectively sample a feature space that is far too large for exhaustive enumeration. We demonstrate the power of the concept by applying it to prediction of protein enzymatic activity.     

Geometry preserving projections algorithm for predicting membrane protein types

Given a new uncharacterized protein sequence, a biologist may want to know whether it is a membrane protein or not? If it is, which membrane protein type it belongs to? Knowing the type of an uncharacterized membrane protein often provides useful clues for finding the biological function of the query protein, developing the computational methods to address these questions can be really helpful. In this study, a sequence encoding scheme based on combing pseudo position-specific score matrix (PsePSSM) and dipeptide composition (DC) is introduced to represent protein samples. However, this sequence encoding scheme would correspond to a very high dimensional feature vector. A dimensionality reduction algorithm, the so-called geometry preserving projections (GPP) is introduced to extract the key features from the high-dimensional space and reduce the original high-dimensional vector to a lower-dimensional one. Finally, the K-nearest neighbor (K-NN) and support vector machine (SVM) classifiers are employed to identify the types of membrane proteins based on their reduced low-dimensional features. Our jackknife and independent dataset test results thus obtained are quite encouraging, which indicate that the above methods are used effectively to deal with this complicated problem of predicting the membrane protein type.     

Cross-validation of protein structural class prediction using statistical clustering and neural networks.

We present an approach to predicting protein structural class that uses amino acid composition and hydrophobic pattern frequency information as input to two types of neural networks: (1) a three-layer back-propagation network and (2) a learning vector quantization network. The results of these methods are compared to those obtained from a modified Euclidean statistical clustering algorithm. The protein sequence data used to drive these algorithms consist of the normalized frequency of up to 20 amino acid types and six hydrophobic amino acid patterns. From these frequency values the structural class predictions for each protein (all-alpha, all-beta, or alpha-beta classes) are derived. Examples consisting of 64 previously classified proteins were randomly divided into multiple training (56 proteins) and test (8 proteins) sets. The best performing algorithm on the test sets was the learning vector quantization network using 17 inputs, obtaining a prediction accuracy of 80.2%. The Matthews correlation coefficients are statistically significant for all algorithms and all structural classes. The differences between algorithms are in general not statistically significant. These results show that information exists in protein primary sequences that is easily obtainable and useful for the prediction of protein structural class by neural networks as well as by standard statistical clustering algorithms.     

An efficient method for protein function annotation based on multilayer protein networks

Background

Accurate annotation of protein functions is still a big challenge for understanding life in the post-genomic era. Many computational methods based on protein-protein interaction (PPI) networks have been proposed to predict the function of proteins. However, the precision of these predictions still needs to be improved, due to the incompletion and noise in PPI networks. Integrating network topology and biological information could improve the accuracy of protein function prediction and may also lead to the discovery of multiple interaction types between proteins. Current algorithms generate a single network, which is archived using a weighted sum of all types of protein interactions.

Method

The influences of different types of interactions on the prediction of protein functions are not the same. To address this, we construct multilayer protein networks (MPN) by integrating PPI networks, the domain of proteins, and information on protein complexes. In the MPN, there is more than one type of connections between pairwise proteins. Different types of connections reflect different roles and importance in protein function prediction. Based on the MPN, we propose a new protein function prediction method, named function prediction based on multilayer protein networks (FP-MPN). Given an un-annotated protein, the FP-MPN method visits each layer of the MPN in turn and generates a set of candidate neighbors with known functions. A set of predicted functions for the testing protein is then formed and all of these functions are scored and sorted. Each layer plays different importance on the prediction of protein functions. A number of top-ranking functions are selected to annotate the unknown protein.

Conclusions

The method proposed in this paper was a better predictor when used on Saccharomyces cerevisiae protein data than other function prediction methods previously used. The proposed FP-MPN method takes different roles of connections in protein function prediction into account to reduce the artificial noise by introducing biological information.

     

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...