Support vector machines with selective kernel scaling for protein classification and identification of key amino acid positions

MOTIVATION: Data that characterize primary and tertiary structures of proteins are now accumulating at a rapid and accelerating rate and require automated computational tools to extract critical information relating amino acid changes with the spectrum of functionally attributes exhibited by a protein. We propose that immunoglobulin-type beta-domains, which are found in approximate 400 functionally distinct forms in humans alone, provide the immense genetic variation within limited conformational changes that might facilitate the development of new computational tools. As an initial step, we describe here an approach based on Support Vector Machine (SVM) technology to identify amino acid variations that contribute to the functional attribute of pathological self-assembly by some human antibody light chains produced during plasma cell diseases. RESULTS: We demonstrate that SVMs with selective kernel scaling are an effective tool in discriminating between benign and pathologic human immunoglobulin light chains. Initial results compare favorably against manual classification performed by experts and indicate the capability of SVMs to capture the underlying structure of the data. The data set consists of 70 proteins of human antibody kappa1 light chains, each represented by aligned sequences of 120 amino acids. We perform feature selection based on a first-order adaptive scaling algorithm, which confirms the importance of changes in certain amino acid positions and identifies other positions that are key in the characterization of protein function.     

Remote protein homology detection using recurrence quantification analysis and amino acid physicochemical properties

Remote homology detection refers to the detection of structure homology in evolutionarily related proteins with low sequence similarity. Supervised learning algorithms such as support vector machine (SVM) are currently the most accurate methods. In most of these SVM-based methods, efforts have been dedicated to developing new kernels to better use the pairwise alignment scores or sequence profiles. Moreover, amino acids’ physicochemical properties are not generally used in the feature representation of protein sequences. In this article, we present a remote homology detection method that incorporates two novel features: (1) a protein's primary sequence is represented using amino acid's physicochemical properties and (2) the similarity between two proteins is measured using recurrence quantification analysis (RQA). An optimization scheme was developed to select different amino acid indices (up to 10 for a protein family) that are best to characterize the given protein family. The selected amino acid indices may enable us to draw better biological explanation of the protein family classification problem than using other alignment-based methods. An SVM-based classifier will then work on the space described by the RQA metrics. The classification scheme is named as SVM-RQA. Experiments at the superfamily level of the SCOP1.53 dataset show that, without using alignment or sequence profile information, the features generated from amino acid indices are able to produce results that are comparable to those obtained by the published state-of-the-art SVM kernels. In the future, better prediction accuracies can be expected by combining the alignment-based features with our amino acids property-based features. Supplementary information including the raw dataset, the best-performing amino acid indices for each protein family and the computed RQA metrics for all protein sequences can be downloaded from http://ym151113.ym.edu.tw/svm-rqa.     

G蛋白偶联受体及其类型的预测

G蛋白偶联受体是非常重要的信号分子受体,其功能失调会导致许多疾病的产生。在前期工作的基础上,作者将序列特征分析与支持向量机技术结合起来,通过分析序列的特征差异,对G蛋白偶联受体分子及其类型进行识别。首次提取了G蛋白偶联受体对应的mRNA序列的绝对密码子使用频率作为特征,这主要因为它既包含了基因密码子使用偏性的信息,也包含了基因所编码蛋白的氨基酸组成信息。结果显示：在G蛋白偶联受体序列及其类型预测的问题中,设计支持向量机分类器时,最好选择使用包含基因序列绝对密码子使用频率和蛋白序列双联氨基酸使用频率两部分信息的组合特征作为特征,同时采用径向基核作为核函数。     

Feature Selection and the Class Imbalance Problem in Predicting Protein Function from Sequence

When the standard approach to predict protein function by sequence homology fails, other alternative methods can be used that require only the amino acid sequence for predicting function. One such approach uses machine learning to predict protein function directly from amino acid sequence features. However, there are two issues to consider before successful functional prediction can take place: identifying discriminatory features, and overcoming the challenge of a large imbalance in the training data. We show that by applying feature subset selection followed by undersampling of the majority class, significantly better support vector machine (SVM) classifiers are generated compared with standard machine learning approaches. As well as revealing that the features selected could have the potential to advance our understanding of the relationship between sequence and function, we also show that undersampling to produce fully balanced data significantly improves performance. The best discriminating ability is achieved using SVMs together with feature selection and full undersampling; this approach strongly outperforms other competitive learning algorithms. We conclude that this combined approach can generate powerful machine learning classifiers for predicting protein function directly from sequence.     

An eco-informatics tool for microbial community studies: supervised classification of Amplicon Length Heterogeneity (ALH) profiles of 16S rRNA

Support vector machines (SVM) and K-nearest neighbors (KNN) are two computational machine learning tools that perform supervised classification. This paper presents a novel application of such supervised analytical tools for microbial community profiling and to distinguish patterning among ecosystems. Amplicon length heterogeneity (ALH) profiles from several hypervariable regions of 16S rRNA gene of eubacterial communities from Idaho agricultural soil samples and from Chesapeake Bay marsh sediments were separately analyzed. The profiles from all available hypervariable regions were concatenated to obtain a combined profile, which was then provided to the SVM and KNN classifiers. Each profile was labeled with information about the location or time of its sampling. We hypothesized that after a learning phase using feature vectors from labeled ALH profiles, both these classifiers would have the capacity to predict the labels of previously unseen samples. The resulting classifiers were able to predict the labels of the Idaho soil samples with high accuracy. The classifiers were less accurate for the classification of the Chesapeake Bay sediments suggesting greater similarity within the Bay's microbial community patterns in the sampled sites. The profiles obtained from the V1+V2 region were more informative than that obtained from any other single region. However, combining them with profiles from the V1 region (with or without the profiles from the V3 region) resulted in the most accurate classification of the samples. The addition of profiles from the V 9 region appeared to confound the classifiers. Our results show that SVM and KNN classifiers can be effectively applied to distinguish between eubacterial community patterns from different ecosystems based only on their ALH profiles.     

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 and classification of human G-protein coupled receptors based on support vector machines

A computational system for the prediction and classification of human G-protein coupled receptors （GPCRs） has been developed based on the support vector machine （SVM） method and protein sequence information. The feature vectors used to develop the SVM prediction models consist of statistically significant features selected from single amino acid, dipeptide, and tripeptide compositions of protein sequences. Furthermore, the length distribution difference between GPCRs and non-GPCRs has also been exploited to improve the prediction performance. The testing results with annotated human protein sequences demonstrate that this system can get good performance for both prediction and classification of human GPCRs.     

Predicting rRNA-, RNA-, and DNA-binding proteins from primary structure with support vector machines

In the post-genome era, the prediction of protein function is one of the most demanding tasks in the study of bioinformatics. Machine learning methods, such as the support vector machines (SVMs), greatly help to improve the classification of protein function. In this work, we integrated SVMs, protein sequence amino acid composition, and associated physicochemical properties into the study of nucleic-acid-binding proteins prediction. We developed the binary classifications for rRNA-, RNA-, DNA-binding proteins that play an important role in the control of many cell processes. Each SVM predicts whether a protein belongs to rRNA-, RNA-, or DNA-binding protein class. Self-consistency and jackknife tests were performed on the protein data sets in which the sequences identity was < 25%. Test results show that the accuracies of rRNA-, RNA-, DNA-binding SVMs predictions are approximately 84%, approximately 78%, approximately 72%, respectively. The predictions were also performed on the ambiguous and negative data set. The results demonstrate that the predicted scores of proteins in the ambiguous data set by RNA- and DNA-binding SVM models were distributed around zero, while most proteins in the negative data set were predicted as negative scores by all three SVMs. The score distributions agree well with the prior knowledge of those proteins and show the effectiveness of sequence associated physicochemical properties in the protein function prediction. The software is available from the author upon request.     

CyclinPred: a SVM-based method for predicting cyclin protein sequences

Functional annotation of protein sequences with low similarity to well characterized protein sequences is a major challenge of computational biology in the post genomic era. The cyclin protein family is once such important family of proteins which consists of sequences with low sequence similarity making discovery of novel cyclins and establishing orthologous relationships amongst the cyclins, a difficult task. The currently identified cyclin motifs and cyclin associated domains do not represent all of the identified and characterized cyclin sequences. We describe a Support Vector Machine (SVM) based classifier, CyclinPred, which can predict cyclin sequences with high efficiency. The SVM classifier was trained with features of selected cyclin and non cyclin protein sequences. The training features of the protein sequences include amino acid composition, dipeptide composition, secondary structure composition and PSI-BLAST generated Position Specific Scoring Matrix (PSSM) profiles. Results obtained from Leave-One-Out cross validation or jackknife test, self consistency and holdout tests prove that the SVM classifier trained with features of PSSM profile was more accurate than the classifiers based on either of the other features alone or hybrids of these features. A cyclin prediction server--CyclinPred has been setup based on SVM model trained with PSSM profiles. CyclinPred prediction results prove that the method may be used as a cyclin prediction tool, complementing conventional cyclin prediction methods.     

Profile-based string kernels for remote homology detection and motif extraction

We introduce novel profile-based string kernels for use with support vector machines (SVMs) for the problems of protein classification and remote homology detection. These kernels use probabilistic profiles, such as those produced by the PSI-BLAST algorithm, to define position-dependent mutation neighborhoods along protein sequences for inexact matching of k-length subsequences ("k-mers") in the data. By use of an efficient data structure, the kernels are fast to compute once the profiles have been obtained. For example, the time needed to run PSI-BLAST in order to build the profiles is significantly longer than both the kernel computation time and the SVM training time. We present remote homology detection experiments based on the SCOP database where we show that profile-based string kernels used with SVM classifiers strongly outperform all recently presented supervised SVM methods. We further examine how to incorporate predicted secondary structure information into the profile kernel to obtain a small but significant performance improvement. We also show how we can use the learned SVM classifier to extract "discriminative sequence motifs"--short regions of the original profile that contribute almost all the weight of the SVM classification score--and show that these discriminative motifs correspond to meaningful structural features in the protein data. The use of PSI-BLAST profiles can be seen as a semi-supervised learning technique, since PSI-BLAST leverages unlabeled data from a large sequence database to build more informative profiles. Recently presented "cluster kernels" give general semi-supervised methods for improving SVM protein classification performance. We show that our profile kernel results also outperform cluster kernels while providing much better scalability to large datasets.     

TSSub: eukaryotic protein subcellular localization by extracting features from profiles

This paper introduces a new subcellular localization system (TSSub) for eukaryotic proteins. This system extracts features from both profiles and amino acid sequences. Four different features are extracted from profiles by four probabilistic neural network (PNN) classifiers, respectively (the amino acid composition from whole profiles; the amino acid composition from the N-terminus of profiles; the dipeptide composition from whole profiles and the amino acid composition from fragments of profiles). In addition, a support vector machine (SVM) classifier is added to implement the residue-couple feature extracted from amino acid sequences. The results from the five classifiers are fused by an additional SVM classifier. The overall accuracies of this TSSub reach 93.0 and 77.4% on Reinhardt and Hubbard's eukaryotic protein dataset and Huang and Li's eukaryotic protein dataset, respectively. The comparison with existing methods results shows TSSub provides better prediction performance than existing methods. AVAILABILITY: The web server is available from http://166.111.24.5/webtools/TSSub/index.html.     

PSP_MCSVM: brainstorming consensus prediction of protein secondary structures using two-stage multiclass support vector machines

Secondary structure prediction is a crucial task for understanding the variety of protein structures and performed biological functions. Prediction of secondary structures for new proteins using their amino acid sequences is of fundamental importance in bioinformatics. We propose a novel technique to predict protein secondary structures based on position-specific scoring matrices (PSSMs) and physico-chemical properties of amino acids. It is a two stage approach involving multiclass support vector machines (SVMs) as classifiers for three different structural conformations, viz., helix, sheet and coil. In the first stage, PSSMs obtained from PSI-BLAST and five specially selected physicochemical properties of amino acids are fed into SVMs as features for sequence-to-structure prediction. Confidence values for forming helix, sheet and coil that are obtained from the first stage SVM are then used in the second stage SVM for performing structure-to-structure prediction. The two-stage cascaded classifiers (PSP_MCSVM) are trained with proteins from RS126 dataset. The classifiers are finally tested on target proteins of critical assessment of protein structure prediction experiment-9 (CASP9). PSP_MCSVM with brainstorming consensus procedure performs better than the prediction servers like Predator, DSC, SIMPA96, for randomly selected proteins from CASP9 targets. The overall performance is found to be comparable with the current state-of-the art. PSP_MCSVM source code, train-test datasets and supplementary files are available freely in public domain at: and     

MMDB: Entrez’s 3D-structure database

Three-dimensional structures are now known within many protein families and it is quite likely, in searching a sequence database, that one will encounter a homolog with known structure. The goal of Entrez’s 3D-structure database is to make this information, and the functional annotation it can provide, easily accessible to molecular biologists. To this end Entrez’s search engine provides three powerful features. (i) Sequence and structure neighbors; one may select all sequences similar to one of interest, for example, and link to any known 3D structures. (ii) Links between databases; one may search by term matching in MEDLINE, for example, and link to 3D structures reported in these articles. (iii) Sequence and structure visualization; identifying a homolog with known structure, one may view molecular-graphic and alignment displays, to infer approximate 3D structure. In this article we focus on two features of Entrez’s Molecular Modeling Database (MMDB) not described previously: links from individual biopolymer chains within 3D structures to a systematic taxonomy of organisms represented in molecular databases, and links from individual chains (and compact 3D domains within them) to structure neighbors, other chains (and 3D domains) with similar 3D structure. MMDB may be accessed at http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=Structure.     

Primary structure of human fibronectin: differential splicing may generate at least 10 polypeptides from a single gene.

Cellular and plasma fibronectins are heterodimers consisting of similar but not identical polypeptides. The differences between fibronectin subunits are due in part to the variability of internal primary sequences. This results from alternative splicing in at least two regions (ED and IIICS) of the pre-mRNA. The complete primary structure of human fibronectin, including most of the internal variations, has been determined by sequencing a series of overlapping cDNA clones. In total, they covered 7692 nucleotides and represented the mRNA sequence coding from the amino terminus of the mature protein to the poly(A) tail. The deduced amino acid sequence of fibronectin has been analysed in terms of the arrangement of internal homologies and the different binding domains.     

Sequence similarities of protein kinase peptide substrates and inhibitors: comparison of their primary structures with immunoglobulin repeats

Forty original sequences of peptide substrates and inhibitors of protein kinases and phosphatases were aligned in a chain matrix without artificial gaps. Fifteen protein kinase peptide substrates and inhibitors (PKSI peptides) contained a common dipeptide ArgArg and also additional important tetra-, tri- and dipeptide homologies. Three further peptide substrates were significantly similar to these peptides but lacked the ArgArg dipeptide. Sequence comparison of individual PKSI peptides revealed probabilistically restricted consensus sequence—PKSI motif—comprising 8 homologous and 13 non-randomly distributed amino acids without considering mutation analysis. This template motif was compared with the consensus sequences of 12 different immunoglobulin domains. In 11 of 12 these domains, the starts of homologous segments were found at nearly the same domain related sites, beginning with serine. A single-triplet mutation, of any of the first two triplet bases that encode equally localized amino acids in each of the two sequence sets (PKSI and Ig) revealed additional homologies with the other set. A primary derived motif version composed of 9 homologous and seven non-randomly distributed amino acids was consequently established by its feedback projection into the original sequence sets. This procedure yielded a second preliminary motif version (revised motif) formed by a sequence of 9 homologous amino acids and two non-randomly distributed amino acids. In addition, three shorter oligopeptide motifs called important stereotypes were derived, based on repeated homology between Ig chains and the revised motif. The most extensive similarities in terms of these stereotypes occurred in the CH2 and CH4 domains of Ig peptides, and inhibitors of cAMP dependent protein kinase and protein kinase A. Further comparisons based on a reference sequence set arranged with the aid of feedback projection revealed a lower similarity between variable Ig chains reflected in a decreased number of homologous amino acids. Two final motif versions, FMC and FMV, were found in two different subsets of constant and variable Ig chains, respectively. FMC was composed of seven homologous and one non-randomly distributed amino acids forming the dispersed structure STLR(C)LVSD, whereas 6 homologous and one questionable amino acid constituted FMV. Only CH4 and CH1 domain segments contained all five high-incidence amino acids, which represented a higher level of similarity than homologous amino acids of all preliminary and final motifs. Four such amino acids were present also in three PKSI peptides. All similarities described here occur in domain segments positionally overlapping with the CDR1 region of variable chains. The results are discussed in terms of immunoglobulin evolution, the position of Fc receptor binding sites and degeneration or mutability of the triplets of motif-constituting amino acids.     

Nucleotide sequence of a rabbit IgG heavy chain from the recombinant F-I haplotype

We report the sequence of a cDNA encoding a rabbit immunoglobulin gamma heavy chain of d12 and e14 allotypes with high homology to partial cDNA sequences from rabbits of d11 and e15 allotypes. The encoded rabbit protein shows homologies with human (68-70%) and mouse (60-63%) gamma chains. The nucleotide sequence homologies of the CH domains range from 76-84% with human and 64-76% with mouse sequences. Comparison of the portion of VH encoding amino acid positions 34-112 with a previously determined VH sequence of the same allotype shows high conservation of sequences in the second and third framework segments but more marked differences both in length and encoded amino acids of the second and third complementarity-determining regions (CDRs). We also found a high degree of homology with a human genomic V-region, VH26 (77%) and a remarkable similarity between rabbit and human second CDR sequences and human genomic D minigenes. These results provide additional evidence that D minigene sequences share information with the CDR2 portion of VH regions.     

Prediction of flexible/rigid regions from protein sequences using k-spaced amino acid pairs

Background

Traditionally, it is believed that the native structure of a protein corresponds to a global minimum of its free energy. However, with the growing number of known tertiary (3D) protein structures, researchers have discovered that some proteins can alter their structures in response to a change in their surroundings or with the help of other proteins or ligands. Such structural shifts play a crucial role with respect to the protein function. To this end, we propose a machine learning method for the prediction of the flexible/rigid regions of proteins (referred to as FlexRP); the method is based on a novel sequence representation and feature selection. Knowledge of the flexible/rigid regions may provide insights into the protein folding process and the 3D structure prediction.

Results

The flexible/rigid regions were defined based on a dataset, which includes protein sequences that have multiple experimental structures, and which was previously used to study the structural conservation of proteins. Sequences drawn from this dataset were represented based on feature sets that were proposed in prior research, such as PSI-BLAST profiles, composition vector and binary sequence encoding, and a newly proposed representation based on frequencies of k-spaced amino acid pairs. These representations were processed by feature selection to reduce the dimensionality. Several machine learning methods for the prediction of flexible/rigid regions and two recently proposed methods for the prediction of conformational changes and unstructured regions were compared with the proposed method. The FlexRP method, which applies Logistic Regression and collocation-based representation with 95 features, obtained 79.5% accuracy. The two runner-up methods, which apply the same sequence representation and Support Vector Machines (SVM) and Naïve Bayes classifiers, obtained 79.2% and 78.4% accuracy, respectively. The remaining considered methods are characterized by accuracies below 70%. Finally, the Naïve Bayes method is shown to provide the highest sensitivity for the prediction of flexible regions, while FlexRP and SVM give the highest sensitivity for rigid regions.

Conclusion

A new sequence representation that uses k-spaced amino acid pairs is shown to be the most efficient in the prediction of the flexible/rigid regions of protein sequences. The proposed FlexRP method provides the highest prediction accuracy of about 80%. The experimental tests show that the FlexRP and SVM methods achieved high overall accuracy and the highest sensitivity for rigid regions, while the best quality of the predictions for flexible regions is achieved by the Naïve Bayes method.