Protein Classification Problem Research Articles

GENERATING FUZZY RULES FOR PROTEIN CLASSIFICATION

This paper considers the generation of some interpretable fuzzy rules for assigning an amino acid sequence into the appropriate protein superfamily. Since the main objective of this classifier is the interpretability of rules, we have used the distribution of amino acids in the sequences of proteins as features. These features are the occurrence probabilities of six exchange groups in the sequences. To generate the fuzzy rules, we have used some modified versions of a common approach. The generated rules are simple and understandable, especially for biologists. To evaluate our fuzzy classifiers, we have used four protein superfamilies from UniProt database. Experimental results show the comprehensibility of generated fuzzy rules with comparable classification accuracy. Bioinformatics(4)is basically conceptualizing biology in terms of macromolecules and applying informatics techniques to understand and organize the information associated with these molecules. It deals primarily with the application of computer and statistical techniques to the management of biological information. Because of the Human Genome Project and other similar efforts, a large number of biological data are regularly collected. It is important to organize and annotate this massive amount of sequential data to maximize its utility. In this regard, DNA sequences are translated into protein sequences using standard bioinformatics tools. Among these is protein sequence classification, which determines the type or group of proteins to which an unknown protein sequence belongs. One of the benefits from this type of category grouping is that molecular analysis can be carried out within a particular superfamily instead of an individual protein sequence. A protein superfamily consists of protein sequence members that are evolutionally related and therefore functionally and structurally relevant to each other. Several approaches dealing with the protein classification problem have been proposed in the past. These include alignment of protein sequences (2), hidden Markov modeling (14), application of artificial neural networks (23, 24, 25), using

Multi-class Protein Classification Using Adaptive Codes

Predicting a protein's structural class from its amino acid sequence is a fundamental problem in computational biology. Recent machine learning work in this domain has focused on developing new input space representations for protein sequences, that is, string kernels, some of which give state-of-the-art performance for the binary prediction task of discriminating between one class and all the others. However, the underlying protein classification problem is in fact a huge multi-class problem, with over 1000 protein folds and even more structural subcategories organized into a hierarchy. To handle this challenging many-class problem while taking advantage of progress on the binary problem, we introduce an adaptive code approach in the output space of one-vs-the-rest prediction scores. Specifically, we use a ranking perceptron algorithm to learn a weighting of binary classifiers that improves multi-class prediction with respect to a fixed set of output codes. We use a cross-validation set-up to generate output vectors for training, and we define codes that capture information about the protein structural hierarchy. Our code weighting approach significantly improves on the standard one-vs-all method for two difficult multi-class protein classification problems: remote homology detection and fold recognition. Our algorithm also outperforms a previous code learning approach due to Crammer and Singer, trained here using a perceptron, when the dimension of the code vectors is high and the number of classes is large. Finally, we compare against PSI-BLAST, one of the most widely used methods in protein sequence analysis, and find that our method strongly outperforms it on every structure classification problem that we consider. Supplementary data and source code are available at http://www.cs.columbia.edu/compbio/adaptive .

SVM-Fold: a tool for discriminative multi-class protein fold and superfamily recognition

BackgroundPredicting a protein's structural class from its amino acid sequence is a fundamental problem in computational biology. Much recent work has focused on developing new representations for protein sequences, called string kernels, for use with support vector machine (SVM) classifiers. However, while some of these approaches exhibit state-of-the-art performance at the binary protein classification problem, i.e. discriminating between a particular protein class and all other classes, few of these studies have addressed the real problem of multi-class superfamily or fold recognition. Moreover, there are only limited software tools and systems for SVM-based protein classification available to the bioinformatics community.ResultsWe present a new multi-class SVM-based protein fold and superfamily recognition system and web server called SVM-Fold, which can be found at . Our system uses an efficient implementation of a state-of-the-art string kernel for sequence profiles, called the profile kernel, where the underlying feature representation is a histogram of inexact matching k-mer frequencies. We also employ a novel machine learning approach to solve the difficult multi-class problem of classifying a sequence of amino acids into one of many known protein structural classes. Binary one-vs-the-rest SVM classifiers that are trained to recognize individual structural classes yield prediction scores that are not comparable, so that standard "one-vs-all" classification fails to perform well. Moreover, SVMs for classes at different levels of the protein structural hierarchy may make useful predictions, but one-vs-all does not try to combine these multiple predictions. To deal with these problems, our method learns relative weights between one-vs-the-rest classifiers and encodes information about the protein structural hierarchy for multi-class prediction. In large-scale benchmark results based on the SCOP database, our code weighting approach significantly improves on the standard one-vs-all method for both the superfamily and fold prediction in the remote homology setting and on the fold recognition problem. Moreover, our code weight learning algorithm strongly outperforms nearest-neighbor methods based on PSI-BLAST in terms of prediction accuracy on every structure classification problem we consider.ConclusionBy combining state-of-the-art SVM kernel methods with a novel multi-class algorithm, the SVM-Fold system delivers efficient and accurate protein fold and superfamily recognition.

A Lazy Data Mining Approach for Protein Classification

In this work, we propose a new computational technique to solve the protein classification problem. The goal is to predict the functional family of novel protein sequences based on their motif composition. In order to improve the results obtained with other known approaches, we propose a new data mining technique for protein classification based on Bayes' theorem, called highest subset probability (HiSP). To evaluate our proposal, datasets extracted from Prosite, a curated protein family database, are used as experimental datasets. The computational results have shown that the proposed method outperforms other known methods for all tested datasets and looks very promising for problems with characteristics similar to the problem addressed here. In addition, our experiments suggest that HiSP performs well on highly imbalanced datasets.

A simple method for protein structural classification

Since the concept of structural classes of proteins was proposed, the problem of protein classification has been tackled by many groups. Most of their classification criteria are based only on the helix/strand contents of proteins. In this paper, we proposed a method for protein structural classification based on their secondary structure sequences. It is a classification scheme that can confirm existing classifications. Here a mathematical model is constructed to describe protein secondary structure sequences, in which each protein secondary structure sequence corresponds to a transition probability matrix that characterizes and differentiates protein structure numerically. Its application to a set of real data has indicated that our method can classify protein structures correctly. The final classification result is shown schematically. So it is visual to observe the structural classifications, which is different from traditional methods.

A Protein Classification Method Based on Latent Semantic Analysis*.

In this paper a new method that uses Latent Semantic Analysis (LSA) to denote a protein sequence is proposed for researching the protein classification problem. A protein is vectorized according to its content of biological words: patterns and motifs, which are generated by utilizing TEIRESIAS algorithm and MEME/MAST system respectively. More precise description vectors of proteins are obtained through employing LSA. Those vectors are used to classify proteins combined with the Support Vector Machine (SVM). Experiments of family-level protein classification on Structural Classification of Proteins database show that the performance of this method is better than that of the other state-of-the-arts methods.

On Prediction Using Variable Order Markov Models

This paper is concerned with algorithms for prediction of discrete sequences over a finite alphabet, using variable order Markov models. The class of such algorithms is large and in principle includes any lossless compression algorithm. We focus on six prominent prediction algorithms, including Context Tree Weighting (CTW), Prediction by Partial Match (PPM) and Probabilistic Suffix Trees (PSTs). We discuss the properties of these algorithms and compare their performance using real life sequences from three domains: proteins, English text and music pieces. The comparison is made with respect to prediction quality as measured by the average log-loss. We also compare classification algorithms based on these predictors with respect to a number of large protein classification tasks. Our results indicate that a ``decomposed'' CTW (a variant of the CTW algorithm) and PPM outperform all other algorithms in sequence prediction tasks. Somewhat surprisingly, a different algorithm, which is a modification of the Lempel-Ziv compression algorithm, significantly outperforms all algorithms on the protein classification problems.

Mismatch string kernels for discriminative protein classification

Classification of proteins sequences into functional and structural families based on sequence homology is a central problem in computational biology. Discriminative supervised machine learning approaches provide good performance, but simplicity and computational efficiency of training and prediction are also important concerns. We introduce a class of string kernels, called mismatch kernels, for use with support vector machines (SVMs) in a discriminative approach to the problem of protein classification and remote homology detection. These kernels measure sequence similarity based on shared occurrences of fixed-length patterns in the data, allowing for mutations between patterns. Thus, the kernels provide a biologically well-motivated way to compare protein sequences without relying on family-based generative models such as hidden Markov models. We compute the kernels efficiently using a mismatch tree data structure, allowing us to calculate the contributions of all patterns occurring in the data in one pass while traversing the tree. When used with an SVM, the kernels enable fast prediction on test sequences. We report experiments on two benchmark SCOP datasets, where we show that the mismatch kernel used with an SVM classifier performs competitively with state-of-the-art methods for homology detection, particularly when very few training examples are available. Examination of the highest-weighted patterns learned by the SVM classifier recovers biologically important motifs in protein families and superfamilies.

Protein Family Classification Using Second-Order Recurrent Neural Networks

Classification of protein sequences into functional and structural families based on sequence homology is a central problem in computational biology. Existing methods for protein classification, including BLAST [1] and pattern/profile search methods[2, 5, 6], rely on finding similarity between a query sequence and protein sequences in databases with known functions. They have difficulties in classifying protein family sharing very weak similarity. And pattern/profile methods rely on multiple alignments for generating profiles or derive patterns from short regions which allows only limited flexibility. In this paper, we formulate protein family classification as a formal language problem; subsequently we give an approach based on recurrent neural networks to classify protein sequences. We define an amino acid sequence to be a tuple X = (X1, X2, . . . , XT ) where Xt ∈ ΣR denotes the tth element in the amino acid sequence and ΣR denotes the set of 20 amino acids. All amino acid sequences can be treated as the language Lprotein from amino acid alphabet ΣR and one family of proteins as a subset of Lprotein, represented by Lf where f denotes the family. So the task is to simulate the function fLf : Lprotein → {accept, reject} which is able to determine whether a given protein sequence X belongs to a protein family f . Learning the discrimination function fLf from examples of the language Lf is a special form of grammatical inference. Symbolic algorithms in this area have not led to successful protein classification because of their lack of robustness in their presence of noisy data. The main connectionist architecture that has been investigated for grammatical inference is recurrent neural network[3, 4]. To solve the problem of protein classification, we propose a second-order recurrent neural network, based on simulating conditional probabilities.