Prediction of protein structural classes by modified mahalanobis discriminant algorithm

Construction of Discrimination Models of Cationic Drugs for Phospholipidosis Induction Potential by Using Interaction Data with Immobilized Artificial Membrane as Well as Physicochemical Properties

Accurate prediction of the phospholipidosis-induction risk of drugs at early stages is important in drug development. So far, discrimination models for predicting the induction risk of cationic drugs have been proposed, but it is still challenging to accurately predict the risk of cationic drugs with intermediate hydrophobicity (logP). In this study, we introduced a parameter (Δlogk40) reflecting not only hydrophobic interaction but also interactions with the polar headgroup between cationic drugs and phospholipids, obtained with liquid chromatography using an immobilized artificial membrane column. The parameter was used along with other physicochemical properties as features to construct discrimination models. Linear discriminant analysis, the modified Mahalanobis discriminant analysis, support vector machine, and random forest were employed for model construction. The results showed that all discrimination models exhibited good predictive performance, with the modified Mahalanobis discriminant analysis and random forest providing the best results for cationic drugs, suggesting that the usefulness of the parameter reflecting complex interactions between cationic drugs and immobilized artificial membrane for constructing discrimination models to predict the induction risk. Furthermore, by applying the parameter as a feature in constructing discrimination models, we demonstrated an improvement in the predictive performance for drugs with intermediate hydrophobicity.

Recent Advances in the Prediction of Protein Structural Classes: Feature Descriptors and Machine Learning Algorithms

In the postgenomic age, rapid growth in the number of sequence-known proteins has been accompanied by much slower growth in the number of structure-known proteins (as a result of experimental limitations), and a widening gap between the two is evident. Because protein function is linked to protein structure, successful prediction of protein structure is of significant importance in protein function identification. Foreknowledge of protein structural class can help improve protein structure prediction with significant medical and pharmaceutical implications. Thus, a fast, suitable, reliable, and reasonable computational method for protein structural class prediction has become pivotal in bioinformatics. Here, we review recent efforts in protein structural class prediction from protein sequence, with particular attention paid to new feature descriptors, which extract information from protein sequence, and the use of machine learning algorithms in both feature selection and the construction of new classification models. These new feature descriptors include amino acid composition, sequence order, physicochemical properties, multiprofile Bayes, and secondary structure-based features. Machine learning methods, such as artificial neural networks (ANNs), support vector machine (SVM), K-nearest neighbor (KNN), random forest, deep learning, and examples of their application are discussed in detail. We also present our view on possible future directions, challenges, and opportunities for the applications of machine learning algorithms for prediction of protein structural classes.

An Insightful 10-year Recollection Since the Emergence of the 5-steps Rule

OBJECTIVE

One of the most challenging and also the most difficult problems is how to formulate a biological sequence with a vector but considerably keep its sequence order information.

METHODS

To address such a problem, the approach of Pseudo Amino Acid Components or PseAAC has been developed.

RESULTS AND CONCLUSION

It has become increasingly clear via the 10-year recollection that the aforementioned proposal has been indeed very powerful.

nifPred: Proteome-Wide Identification and Categorization of Nitrogen-Fixation Proteins of Diaztrophs Based on Composition-Transition-Distribution Features Using Support Vector Machine

As inorganic nitrogen compounds are essential for basic building blocks of life (e.g., nucleotides and amino acids), the role of biological nitrogen-fixation (BNF) is indispensible. All nitrogen fixing microbes rely on the same nitrogenase enzyme for nitrogen reduction, which is in fact an enzyme complex consists of as many as 20 genes. However, the occurrence of six genes viz., nifB, nifD, nifE, nifH, nifK, and nifN has been proposed to be essential for a functional nitrogenase enzyme. Therefore, identification of these genes is important to understand the mechanism of BNF as well as to explore the possibilities for improving BNF from agricultural sustainability point of view. Further, though the computational tools are available for the annotation and phylogenetic analysis of nifH gene sequences alone, to the best of our knowledge no tool is available for the computational prediction of the above mentioned six categories of nitrogen-fixation (nif) genes or proteins. Thus, we proposed an approach, which is first of its kind for the computational identification of nif proteins encoded by the six categories of nif genes. Sequence-derived features were employed to map the input sequences into vectors of numeric observations that were subsequently fed to the support vector machine as input. Two types of classifier were constructed: (i) a binary classifier for classification of nif and non-nitrogen-fixation (non-nif) proteins, and (ii) a multi-class classifier for classification of six categories of nif proteins. Higher accuracies were observed for the combination of composition-transition-distribution (CTD) feature set and radial kernel, as compared to the other feature-kernel combinations. The overall accuracies were observed >90% in both binary and multi-class classifications. The developed approach further achieved >92% accuracy, while evaluated with blind (independent) test datasets. The developed approach also produced higher accuracy in identifying nif proteins, while evaluated using proteome-wide datasets of several species. Furthermore, we established a prediction server nifPred (http://webapp.cabgrid.res.in/nifPred) to assist the scientific community for proteome-wide identification of six categories of nif proteins. Besides, the source code of nifPred is also available at https://github.com/PrabinaMeher/nifPred. The developed web server is expected to supplement the transcriptional profiling and comparative genomics studies for the identification and functional annotation of genes related to BNF.

iNuc-PhysChem: a sequence-based predictor for identifying nucleosomes via physicochemical properties

Nucleosome positioning has important roles in key cellular processes. Although intensive efforts have been made in this area, the rules defining nucleosome positioning is still elusive and debated. In this study, we carried out a systematic comparison among the profiles of twelve DNA physicochemical features between the nucleosomal and linker sequences in the Saccharomyces cerevisiae genome. We found that nucleosomal sequences have some position-specific physicochemical features, which can be used for in-depth studying nucleosomes. Meanwhile, a new predictor, called iNuc-PhysChem, was developed for identification of nucleosomal sequences by incorporating these physicochemical properties into a 1788-D (dimensional) feature vector, which was further reduced to a 884-D vector via the IFS (incremental feature selection) procedure to optimize the feature set. It was observed by a cross-validation test on a benchmark dataset that the overall success rate achieved by iNuc-PhysChem was over 96% in identifying nucleosomal or linker sequences. As a web-server, iNuc-PhysChem is freely accessible to the public at http://lin.uestc.edu.cn/server/iNuc-PhysChem. For the convenience of the vast majority of experimental scientists, a step-by-step guide is provided on how to use the web-server to get the desired results without the need to follow the complicated mathematics that were presented just for the integrity in developing the predictor. Meanwhile, for those who prefer to run predictions in their own computers, the predictor's code can be easily downloaded from the web-server. It is anticipated that iNuc-PhysChem may become a useful high throughput tool for both basic research and drug design.

GPCR-2L: predicting G protein-coupled receptors and their types by hybridizing two different modes of pseudo amino acid compositions

G protein-coupled receptors (GPCRs) are among the most frequent targets of therapeutic drugs. With the avalanche of newly generated protein sequences in the post genomic age, to expedite the process of drug discovery, it is highly desirable to develop an automated method to rapidly identify GPCRs and their types. A new predictor was developed by hybridizing two different modes of pseudo-amino acid composition (PseAAC): the functional domain PseAAC and the low-frequency Fourier spectrum PseAAC. The new predictor is called GPCR-2L, where "2L" means that it is a two-layer predictor: the 1st layer prediction engine is to identify a query protein as GPCR or not; if it is, the prediction will be automatically continued to further identify it as belonging to one of the following six types: (1) rhodopsin-like (Class A), (2) secretin-like (Class B), (3) metabotropic glutamate/pheromone (Class C), (4) fungal pheromone (Class D), (5) cAMP receptor (Class E), or (6) frizzled/smoothened family (Class F). The overall success rate of GPCR-2L in identifying proteins as GPCRs or non-GPCRs is over 97.2%, while identifying GPCRs among their six types is over 97.8%. Such high success rates were derived by the rigorous jackknife cross-validation on a stringent benchmark dataset, in which none of the included proteins had ≥40% pairwise sequence identity to any other protein in a same subset. As a user-friendly web-server, GPCR-2L is freely accessible to the public at http://icpr.jci.edu.cn/, by which one can obtain the 2-level results in about 20 s for a query protein sequence of 500 amino acids. The longer the sequence is, the more time it may usually need. The high success rates reported here indicate that it is a quite effective approach to identify GPCRs and their types with the functional domain information and the low-frequency Fourier spectrum analysis. It is anticipated that GPCR-2L may become a useful tool for both basic research and drug development in the areas related to GPCRs.

Some remarks on protein attribute prediction and pseudo amino acid composition

With the accomplishment of human genome sequencing, the number of sequence-known proteins has increased explosively. In contrast, the pace is much slower in determining their biological attributes. As a consequence, the gap between sequence-known proteins and attribute-known proteins has become increasingly large. The unbalanced situation, which has critically limited our ability to timely utilize the newly discovered proteins for basic research and drug development, has called for developing computational methods or high-throughput automated tools for fast and reliably identifying various attributes of uncharacterized proteins based on their sequence information alone. Actually, during the last two decades or so, many methods in this regard have been established in hope to bridge such a gap. In the course of developing these methods, the following things were often needed to consider: (1) benchmark dataset construction, (2) protein sample formulation, (3) operating algorithm (or engine), (4) anticipated accuracy, and (5) web-server establishment. In this review, we are to discuss each of the five procedures, with a special focus on the introduction of pseudo amino acid composition (PseAAC), its different modes and applications as well as its recent development, particularly in how to use the general formulation of PseAAC to reflect the core and essential features that are deeply hidden in complicated protein sequences.

Predicting protein structural classes with pseudo amino acid composition: an approach using geometric moments of cellular automaton image

A novel approach was developed for predicting the structural classes of proteins based on their sequences. It was assumed that proteins belonging to the same structural class must bear some sort of similar texture on the images generated by the cellular automaton evolving rule [Wolfram, S., 1984. Cellular automation as models of complexity. Nature 311, 419-424]. Based on this, two geometric invariant moment factors derived from the image functions were used as the pseudo amino acid components [Chou, K.C., 2001. Prediction of protein cellular attributes using pseudo amino acid composition. Proteins: Struct., Funct., Genet. (Erratum: ibid., 2001, vol. 44, 60) 43, 246-255] to formulate the protein samples for statistical prediction. The success rates thus obtained on a previously constructed benchmark dataset are quite promising, implying that the cellular automaton image can help to reveal some inherent and subtle features deeply hidden in a pile of long and complicated amino acid sequences.

Using pseudo amino acid composition to predict protein subnuclear location with improved hybrid approach

The subnuclear localization of nuclear protein is very important for in-depth understanding of the construction and function of the nucleus. Based on the amino acid and pseudo amino acid composition (PseAA) as originally introduced by K. C. Chou can incorporate much more information of a protein sequence than the classical amino acid composition so as to significantly enhance the power of using a discrete model to predict various attributes of a protein, an algorithm of increment of diversity combined with the improved quadratic discriminant analysis is proposed to predict the protein subnuclear location. The overall predictive success rates and correlation coefficient are 75.4% and 0.629 for 504 single localization proteins in jackknife test, and 80.4% for an independent set of 92 multi-localization proteins, respectively. For 406 single localization nuclear proteins with < or =25% sequence identity, the results of jackknife test show that the overall accuracy of prediction is 77.1%.

Large-scale plant protein subcellular location prediction

Current plant genome sequencing projects have called for development of novel and powerful high throughput tools for timely annotating the subcellular location of uncharacterized plant proteins. In view of this, an ensemble classifier, Plant-PLoc, formed by fusing many basic individual classifiers, has been developed for large-scale subcellular location prediction for plant proteins. Each of the basic classifiers was engineered by the K-Nearest Neighbor (KNN) rule. Plant-PLoc discriminates plant proteins among the following 11 subcellular locations: (1) cell wall, (2) chloroplast, (3) cytoplasm, (4) endoplasmic reticulum, (5) extracell, (6) mitochondrion, (7) nucleus, (8) peroxisome, (9) plasma membrane, (10) plastid, and (11) vacuole. As a demonstration, predictions were performed on a stringent benchmark dataset in which none of the proteins included has > or =25% sequence identity to any other in a same subcellular location to avoid the homology bias. The overall success rate thus obtained was 32-51% higher than the rates obtained by the previous methods on the same benchmark dataset. The essence of Plant-PLoc in enhancing the prediction quality and its significance in biological applications are discussed. Plant-PLoc is accessible to public as a free web-server at: (http://202.120.37.186/bioinf/plant). Furthermore, for public convenience, results predicted by Plant-PLoc have been provided in a downloadable file at the same website for all plant protein entries in the Swiss-Prot database that do not have subcellular location annotations, or are annotated as being uncertain. The large-scale results will be updated twice a year to include new entries of plant proteins and reflect the continuous development of Plant-PLoc.

Using pseudo amino acid composition to predict protein structural class: Approached by incorporating 400 dipeptide components

The proteins structure can be mainly classified into four classes: all-alpha, all-beta, alpha/beta, and alpha + beta protein according to their chain fold topologies. For the purpose of predicting the protein structural class, a new predicting algorithm, in which the increment of diversity combines with Quadratic Discriminant analysis, is presented to study and predict protein structural class. On the basis of the concept of the pseudo amino acid composition (Chou, Proteins: Struct Funct Genet 2001, 43, 246; Erratum: Proteins Struct Funct Genet 2001, 44, 60), 400 dipeptide components and 20 amino acid composition are, respectively, selected as parameters of diversity source. Total of 204 nonhomologous proteins constructed by Chou (Chou, Biochem Biophys Res Commun 1999, 264, 216) are used for training and testing the predictive model. The predicted results by using the pseudo amino acids approach as proposed in this paper can remarkably improve the success rates, and hence the current method may play a complementary role to other existing methods for predicting protein structural classification.

Using pseudo amino acid composition and binary-tree support vector machines to predict protein structural classes

Compared with the conventional amino acid composition (AA), the pseudo amino acid composition (PseAA) as originally introduced by Chou can incorporate much more information of a protein sequence; this remarkably enhances the power to use a discrete model for predicting various attributes of a protein. In this study, based on the concept of Chou's PseAA, a 46-D (dimensional) PseAA was formulated to represent the sample of a protein and a new approach based on binary-tree support vector machines (BTSVMs) was proposed to predict the protein structural class. BTSVMs algorithm has the capability in solving the problem of unclassifiable data points in multi-class SVMs. The results by both the 10-fold cross-validation and jackknife tests demonstrate that the predictive performance using the new PseAA (46-D) is better than that of AA (20-D), which is widely used in many algorithms for protein structural class prediction. The results obtained by the new approach are quite encouraging, indicating that it can at least play a complimentary role to many of the existing methods and is a useful tool for predicting many other protein attributes as well.

Predicting conotoxin superfamily and family by using pseudo amino acid composition and modified Mahalanobis discriminant

The conotoxin proteins are disulfide rich small peptides that target ion channels and G protein coupled receptors. And they provide promising application in treating some chronic pain, epilepsy, cardiovascular diseases, and so on. Conotoxins may be classified into 11 superfamilies: A, D, I1, I2, J, L, M, O, P, S, and T according to the disulfide connectivity, highly conserved N-terminal precursor sequence and similar mode of actions. Successful prediction mature conotoxin superfamily peptide has important signification for the biological and pharmacological functions of the toxins. In this study, a new algorithm of increment of diversity combined with modified Mahalanobis discriminant is presented to predict five superfamilies by using the pseudo amino acid composition. The results of jackknife cross-validation test show that the overall prediction sensitivity and specificity are 88% and 91%, respectively. The predictive algorithm is also used to predict three O-conotoxin families. The 72% sensitivity and 78% specificity are obtained. These results indicate that the conotoxin superfamily peptides correlate with their amino acid compositions.

Amino Acid Principal Component Analysis (AAPCA) and its applications in protein structural class prediction

The extremely complicated nature of many biological problems makes them bear the features of fuzzy sets, such as with vague, imprecise, noisy, ambiguous, or input-missing information For instance, the current data in classifying protein structural classes are typically a fuzzy set To deal with this kind of problem, the AAPCA (Amino Acid Principal Component Analysis) approach was introduced. In the AAPCA approach the 20-dimensional amino acid composition space is reduced to an orthogonal space with fewer dimensions, and the original base functions are converted into a set of orthogonal and normalized base functions The advantage of such an approach is that it can minimize the random errors and redundant information in protein dataset through a principal component selection, remarkably improving the success rates in predicting protein structural classes It is anticipated that the AAPCA approach can be used to deal with many other classification problems in proteins as well.

Using pseudo amino acid composition to predict protein structural classes: approached with complexity measure factor

The structural class is an important feature widely used to characterize the overall folding type of a protein. How to improve the prediction quality for protein structural classification by effectively incorporating the sequence-order effects is an important and challenging problem. Based on the concept of the pseudo amino acid composition [Chou, K. C. Proteins Struct Funct Genet 2001, 43, 246; Erratum: Proteins Struct Funct Genet 2001, 44, 60], a novel approach for measuring the complexity of a protein sequence was introduced. The advantage by incorporating the complexity measure factor into the pseudo amino acid composition as one of its components is that it can catch the essence of the overall sequence pattern of a protein and hence more effectively reflect its sequence-order effects. It was demonstrated thru the jackknife crossvalidation test that the overall success rate by the new approach was significantly higher than those by the others. It has not escaped our notice that the introduction of the complexity measure factor can also be used to improve the prediction quality for, among many other protein attributes, subcellular localization, enzyme family class, membrane protein type, and G-protein couple receptor type.

Using LogitBoost classifier to predict protein structural classes

Prediction of protein classification is an important topic in molecular biology. This is because it is able to not only provide useful information from the viewpoint of structure itself, but also greatly stimulate the characterization of many other features of proteins that may be closely correlated with their biological functions. In this paper, the LogitBoost, one of the boosting algorithms developed recently, is introduced for predicting protein structural classes. It performs classification using a regression scheme as the base learner, which can handle multi-class problems and is particularly superior in coping with noisy data. It was demonstrated that the LogitBoost outperformed the support vector machines in predicting the structural classes for a given dataset, indicating that the new classifier is very promising. It is anticipated that the power in predicting protein structural classes as well as many other bio-macromolecular attributes will be further strengthened if the LogitBoost and some other existing algorithms can be effectively complemented with each other.

Using supervised fuzzy clustering to predict protein structural classes

Prediction of protein classification is both an important and a tempting topic in protein science. This is because of not only that the knowledge thus obtained can provide useful information about the overall structure of a query protein, but also that the practice itself can technically stimulate the development of novel predictors that may be straightforwardly applied to many other relevant areas. In this paper, a novel approach, the so-called "supervised fuzzy clustering approach" is introduced that is featured by utilizing the class label information during the training process. Based on such an approach, a set of "if-then" fuzzy rules for predicting the protein structural classes are extracted from a training dataset. It has been demonstrated through two different working datasets that the overall success prediction rates obtained by the supervised fuzzy clustering approach are all higher than those by the unsupervised fuzzy c-means introduced by the previous investigators [C.T. Zhang, K.C. Chou, G.M. Maggiora. Protein Eng. (1995) 8, 425-435]. It is anticipated that the current predictor may play an important complementary role to other existing predictors in this area to further strengthen the power in predicting the structural classes of proteins and their other characteristic attributes.

Using amphiphilic pseudo amino acid composition to predict enzyme subfamily classes

MOTIVATION

With protein sequences entering into databanks at an explosive pace, the early determination of the family or subfamily class for a newly found enzyme molecule becomes important because this is directly related to the detailed information about which specific target it acts on, as well as to its catalytic process and biological function. Unfortunately, it is both time-consuming and costly to do so by experiments alone. In a previous study, the covariant-discriminant algorithm was introduced to identify the 16 subfamily classes of oxidoreductases. Although the results were quite encouraging, the entire prediction process was based on the amino acid composition alone without including any sequence-order information. Therefore, it is worthy of further investigation.

RESULTS

To incorporate the sequence-order effects into the predictor, the 'amphiphilic pseudo amino acid composition' is introduced to represent the statistical sample of a protein. The novel representation contains 20 + 2lambda discrete numbers: the first 20 numbers are the components of the conventional amino acid composition; the next 2lambda numbers are a set of correlation factors that reflect different hydrophobicity and hydrophilicity distribution patterns along a protein chain. Based on such a concept and formulation scheme, a new predictor is developed. It is shown by the self-consistency test, jackknife test and independent dataset tests that the success rates obtained by the new predictor are all significantly higher than those by the previous predictors. The significant enhancement in success rates also implies that the distribution of hydrophobicity and hydrophilicity of the amino acid residues along a protein chain plays a very important role to its structure and function.

A novel approach to predict active sites of enzyme molecules

Enzymes are critical in many cellular signaling cascades. With many enzyme structures being solved, there is an increasing need to develop an automated method for identifying their active sites. However, given the atomic coordinates of an enzyme molecule, how can we predict its active site? This is a vitally important problem because the core of an enzyme molecule is its active site from the viewpoints of both pure scientific research and industrial application. In this article, a topological entity was introduced to characterize the enzymatic active site. Based on such a concept, the covariant discriminant algorithm was formulated for identifying the active site. As a paradigm, the serine hydrolase family was demonstrated. The overall success rate by jackknife test for a data set of 88 enzyme molecules was 99.92%, and that for a data set of 50 independent enzyme molecules was 99.91%. Meanwhile, it was shown through an example that the prediction algorithm can also be used to find any typographic error of a PDB file in annotating the constituent amino acids of catalytic triad and to suggest a possible correction. The very high success rates are due to the introduction of a covariance matrix in the prediction algorithm that makes allowance for taking into account the coupling effects among the key constituent atoms of active site. It is anticipated that the novel approach is quite promising and may become a useful high throughput tool in enzymology, proteomics, and structural bioinformatics.

Prediction and classification of protein subcellular location-sequence-order effect and pseudo amino acid composition

Given a protein sequence, how to identify its subcellular location? With the rapid increase in newly found protein sequences entering into databanks, the problem has become more and more important because the function of a protein is closely correlated with its localization. To practically deal with the challenge, a dataset has been established that allows the identification performed among the following 14 subcellular locations: (1) cell wall, (2) centriole, (3) chloroplast, (4) cytoplasm, (5) cytoskeleton, (6) endoplasmic reticulum, (7) extracellular, (8) Golgi apparatus, (9) lysosome, (10) mitochondria, (11) nucleus, (12) peroxisome, (13) plasma membrane, and (14) vacuole. Compared with the datasets constructed by the previous investigators, the current one represents the largest in the scope of localizations covered, and hence many proteins which were totally out of picture in the previous treatments, can now be investigated. Meanwhile, to enhance the potential and flexibility in taking into account the sequence-order effect, the series-mode pseudo-amino-acid-composition has been introduced as a representation for a protein. High success rates are obtained by the re-substitution test, jackknife test, and independent dataset test, respectively. It is anticipated that the current automated method can be developed to a high throughput tool for practical usage in both basic research and pharmaceutical industry.

Support vector machines for predicting membrane protein types by using functional domain composition

Membrane proteins are generally classified into the following five types: 1), type I membrane protein; 2), type II membrane protein; 3), multipass transmembrane proteins; 4), lipid chain-anchored membrane proteins; and 5), GPI-anchored membrane proteins. In this article, based on the concept of using the functional domain composition to define a protein, the Support Vector Machine algorithm is developed for predicting the membrane protein type. High success rates are obtained by both the self-consistency and jackknife tests. The current approach, complemented with the powerful covariant discriminant algorithm based on the pseudo-amino acid composition that has incorporated quasi-sequence-order effect as recently proposed by K. C. Chou (2001), may become a very useful high-throughput tool in the area of bioinformatics and proteomics.

Application of pseudo amino acid composition for predicting protein subcellular location: stochastic signal processing approach

The function of a protein is closely correlated with its subcellular location. With the success of human genome project and the rapid increase in the number of newly found protein sequences entering into data banks, it is highly desirable to develop an automated method for predicting the subcellular location of proteins. The establishment of such a predictor will no doubt expedite the functionality determination of newly found proteins and the process of prioritizing genes and proteins identified by genomics efforts as potential molecular targets for drug design. Based on the concept of pseudo amino acid composition originally proposed by K. C. Chou (Proteins: Struct. Funct. Genet. 43: 246-255, 2001), the digital signal processing approach has been introduced to partially incorporate the sequence order effect. One of the remarkable merits by doing so is that many existing tools in mathematics and engineering can be straightforwardly used in predicting protein subcellular location. The results thus obtained are quite encouraging. It is anticipated that the digital signal processing may serve as a useful vehicle for many other protein science areas as well.

Subcellular location prediction of apoptosis proteins

Apoptosis proteins have a central role in the development and homeostasis of an organism. These proteins are very important for understanding the mechanism of programmed cell death. Many efforts in pharmaceutical research have been aimed at understanding their structure and function. Unfortunately, thus far, very few apoptosis protein structures have been determined. In contrast, many apoptosis protein sequences are known, and many more are expected to come in the near future. Because of the extremely unbalanced state, it would be worthwhile to develop a fast sequence-based method to identify their subcellular location so as to gain some insight about their biological function. In view of this, a study was initiated in an attempt to identify the subcellular location of apoptosis proteins according to their sequences by means of the covariant discriminant function, which was established based on the Mahalanobis distance and Chou's invariance theorem (Chou, Proteins 1995;21:319-344). The results were quite promising, indicating that the subcellular location of apoptosis proteins are predictable to a considerably accurate extent if a good training data set can be established. It is expected that, with a continuous improvement of the training data set by incorporating more and more new data, the current method might eventually become a useful tool in this area because the function of an apoptosis protein is closely related to its subcellular location.

Using functional domain composition and support vector machines for prediction of protein subcellular location

Proteins are generally classified into the following 12 subcellular locations: 1) chloroplast, 2) cytoplasm, 3) cytoskeleton, 4) endoplasmic reticulum, 5) extracellular, 6) Golgi apparatus, 7) lysosome, 8) mitochondria, 9) nucleus, 10) peroxisome, 11) plasma membrane, and 12) vacuole. Because the function of a protein is closely correlated with its subcellular location, with the rapid increase in new protein sequences entering into databanks, it is vitally important for both basic research and pharmaceutical industry to establish a high throughput tool for predicting protein subcellular location. In this paper, a new concept, the so-called "functional domain composition" is introduced. Based on the novel concept, the representation for a protein can be defined as a vector in a high-dimensional space, where each of the clustered functional domains derived from the protein universe serves as a vector base. With such a novel representation for a protein, the support vector machine (SVM) algorithm is introduced for predicting protein subcellular location. High success rates are obtained by the self-consistency test, jackknife test, and independent dataset test, respectively. The current approach not only can play an important complementary role to the powerful covariant discriminant algorithm based on the pseudo amino acid composition representation (Chou, K. C. (2001) Proteins Struct. Funct. Genet. 43, 246-255; Correction (2001) Proteins Struct. Funct. Genet. 44, 60), but also may greatly stimulate the development of this area.

Prediction of the subcellular location of prokaryotic proteins based on a new representation of the amino acid composition

A new representation of protein sequence is devoted in this paper, in which each protein can be represented by a 20-dimensional (20D) vector of unit length. Inspired by the principle of superposition of state in quantum mechanics, the squares of the 20 components of the vector correspond to the amino acid composition. Using the new representation of the primary sequence and Bayes Discriminant Algorithm, the subcellular location of prokaryotic proteins was predicted. The overall predictive accuracy in the jackknife test can be 3% higher than the result of using amino acid composition directly for the database of sequence identity is less than 90%, but 5% higher when sequence identity is less than 80%. The higher predictive accuracy indicates that the current measure of extracting the information from the primary sequence is efficient. Since the subcellular location restricting a protein's possible function, the present method should also be a useful measure for the systematic analysis of genome data. The program used in this paper is available on request.

Prediction of the subcellular location of prokaryotic proteins based on the hydrophobicity index of amino acids

An algorithm of predicting the subcellular location of prokaryotic proteins is proposed in this paper. In addition to the amino acid composition, the auto-correlation functions based on the hydrophobicity profile of amino acids along the primary sequence of the query protein have been used. Consequently, the best predictive accuracy to date has been achieved. Of the 997 prokaryotic proteins in the database used here, 688 cytoplasmic, 107 extracellular and 202 periplasmic proteins, the overall predictive accuracies are as high as 97.7 and 90.4% in the resubstitution and jackknife tests, respectively, using the hydrophilicity value of Hopp and Woods. The underlying mechanism of the improvement is also discussed. This work would be useful for a systematic analysis of the great amounts of prokaryotic genome sequences. The computer programs used in this paper are available on request via email.

Some insights into protein structural class prediction

It has been quite clear that the success rate for predicting protein structural class can be improved significantly by using the algorithms that incorporate the coupling effect among different amino acid components of a protein. However, there is still a lot of confusion in understanding the relationship of these advanced algorithms, such as the least Mahalanobis distance algorithm, the component-coupled algorithm, and the Bayes decision rule. In this communication, a simple, rigorous derivation is provided to prove that the Bayes decision rule introduced recently for protein structural class prediction is completely the same as the earlier component-coupled algorithm. Meanwhile, it is also very clear from the derivative equations that the least Mahalanobis distance algorithm is an approximation of the component-coupled algorithm, also named as the covariant-discriminant algorithm introduced by Chou and Elrod in protein subcellular location prediction (Protein Engineering, 1999; 12:107-118). Clarification of the confusion will help use these powerful algorithms effectively and correctly interpret the results obtained by them, so as to conduce to the further development not only in the structural prediction area, but in some other relevant areas in protein science as well.

Prediction of tight turns and their types in proteins

A tight turn in protein structure is defined as a site where (i) a polypeptide chain reverses its overall direction, i.e., leads the chain to fold back on itself by nearly 180 degrees, and (ii) the amino acid residues directly involved in forming the turn are no more than six. Tight turns are generally categorized as delta-turn, gamma-turn, beta-turn, alpha-turn, and pi-turn, which are formed by two-, three-, four-, five-, and six-amino-acid residues, respectively. According to the folding mode, each of such tight turns can be further classified into several different types. Tight turns play an important role in globular proteins from both the structural and functional points of view. In view of this, various efforts have been made to predict tight turns and their types. This Review summarizes the development in this area, with an emphasis focused on the most recent work concerned that is featured by the sequence-coupled model. Meanwhile, the future challenge in this area has also been briefly addressed.

Prediction of membrane protein types based on the hydrophobic index of amino acids

A new algorithm to predict the types of membrane proteins is proposed. Besides the amino acid composition of the query protein, the information within the amino acid sequence is taken into account. A formulation of the autocorrelation functions based on the hydrophobicity index of the 20 amino acids is adopted. The overall predictive accuracy is remarkably increased for the database of 2054 membrane proteins studied here. An improvement of about 13% in the resubstitution test and 8% in the jackknife test is achieved compared with those of algorithms based merely on the amino acid composition. Consequently, overall predictive accuracy is as high as 94% and 82% for the resubstitution and jackknife tests, respectively, for the prediction of the five types. Since the proposed algorithm is based on more parameters than those in the amino acid composition approach, the predictive accuracy would be further increased for a larger and more class-balanced database. The present algorithm should be useful in the determination of the types and functions of new membrane proteins. The computer program is available on request.

Prediction of protein (domain) structural classes based on amino-acid index

A protein (domain) is usually classified into one of the following four structural classes: all-alpha, all-beta, alpha/beta and alpha + beta. In this paper, a new formulation is proposed to predict the structural class of a protein (domain) from its primary sequence. Instead of the amino-acid composition used widely in the previous structural class prediction work, the auto-correlation functions based on the profile of amino-acid index along the primary sequence of the query protein (domain) are used for the structural class prediction. Consequently, the overall predictive accuracy is remarkably improved. For the same training database consisting of 359 proteins (domains) and the same component-coupled algorithm [Chou, K.C. & Maggiora, G.M. (1998) Protein Eng. 11, 523-538], the overall predictive accuracy of the new method for the jackknife test is 5-7% higher than the accuracy based only on the amino-acid composition. The overall predictive accuracy finally obtained for the jackknife test is as high as 90.5%, implying that a significant improvement has been achieved by making full use of the information contained in the primary sequence for the class prediction. This improvement depends on the size of the training database, the auto-correlation functions selected and the amino-acid index used. We have found that the amino-acid index proposed by Oobatake and Ooi, i.e. the average nonbonded energy per residue, leads to the optimal predictive result in the case for the database sets studied in this paper. This study may be considered as an alternative step towards making the structural class prediction more practical.

Prediction of protein secondary structure content

All existing algorithms for predicting the content of protein secondary structure elements have been based on the conventional amino-acid-composition, where no sequence coupling effects are taken into account. In this article, an algorithm was developed for predicting the content of protein secondary structure elements that was based on a new amino-acid-composition, in which the sequence coupling effects are explicitly included through a series of conditional probability elements. The prediction was examined by a self-consistency test and an independent dataset test. Both indicated a remarkable improvement obtained when using the current algorithm to predict the contents of alpha-helix, beta-sheet, beta-bridge, 3(10)-helix, pi-helix, H-bonded turn, bend and random coil. Examples of the improved accuracy by introducing the new amino-acid-composition, as well as its impact on the study of protein structural class and biologically function, are discussed.

A key driving force in determination of protein structural classes

The three-dimensional structure of a protein is uniquely dictated by its primary sequence. However, owing to the very high degenerative nature of the sequence-structure relationship, proteins are generally folded into one of only a few structural classes that are closely correlated with the amino-acid composition. This suggests that the interaction among the components of amino acid composition may play a considerable role in determining the structural class of a protein. To quantitatively test such a hypothesis at a deeper level, three potential functions, U((0)), U((1)), and U((2)), were formulated that respectively represent the 0th-order, 1st-order, and 2nd-order approximations for the interaction among the components of the amino acid composition in a protein. It was observed that the correct rates in recognizing protein structural classes by U((2)) are significantly higher than those by U((0)) and U((1)), indicating that an algorithm that can more completely incorporate the interaction contributions will yield better recognition quality, and hence further demonstrate that the interaction among the components of amino acid composition is an important driving force in determining the structural class of a protein during the sequence folding process.

Protein subcellular location prediction

The function of a protein is closely correlated with its subcellular location. With the rapid increase in new protein sequences entering into data banks, we are confronted with a challenge: is it possible to utilize a bioinformatic approach to help expedite the determination of protein subcellular locations? To explore this problem, proteins were classified, according to their subcellular locations, into the following 12 groups: (1) chloroplast, (2) cytoplasm, (3) cytoskeleton, (4) endoplasmic reticulum, (5) extracell, (6) Golgi apparatus, (7) lysosome, (8) mitochondria, (9) nucleus, (10) peroxisome, (11) plasma membrane and (12) vacuole. Based on the classification scheme that has covered almost all the organelles and subcellular compartments in an animal or plant cell, a covariant discriminant algorithm was proposed to predict the subcellular location of a query protein according to its amino acid composition. Results obtained through self-consistency, jackknife and independent dataset tests indicated that the rates of correct prediction by the current algorithm are significantly higher than those by the existing methods. It is anticipated that the classification scheme and concept and also the prediction algorithm can expedite the functionality determination of new proteins, which can also be of use in the prioritization of genes and proteins identified by genomic efforts as potential molecular targets for drug design.