Building multiclass classifiers for remote homology detection and fold recognition

Background  

Protein remote homology detection and fold recognition are central problems in computational biology. Supervised learning algorithms based on support vector machines are currently one of the most effective methods for solving these problems. These methods are primarily used to solve binary classification problems and they have not been extensively used to solve the more general multiclass remote homology prediction and fold recognition problems.     

Profile-based direct kernels for remote homology detection and fold recognition

MOTIVATION: Protein remote homology detection is a central problem in computational biology. Supervised learning algorithms based on support vector machines are currently one of the most effective methods for remote homology detection. The performance of these methods depends on how the protein sequences are modeled and on the method used to compute the kernel function between them. RESULTS: We introduce two classes of kernel functions that are constructed by combining sequence profiles with new and existing approaches for determining the similarity between pairs of protein sequences. These kernels are constructed directly from these explicit protein similarity measures and employ effective profile-to-profile scoring schemes for measuring the similarity between pairs of proteins. Experiments with remote homology detection and fold recognition problems show that these kernels are capable of producing results that are substantially better than those produced by all of the existing state-of-the-art SVM-based methods. In addition, the experiments show that these kernels, even when used in the absence of profiles, produce results that are better than those produced by existing non-profile-based schemes. AVAILABILITY: The programs for computing the various kernel functions are available on request from the authors.     

Scalable remote homology detection and fold recognition in massive protein networks

The global connectivities in very large protein similarity networks contain traces of evolution among the proteins for detecting protein remote evolutionary relations or structural similarities. To investigate how well a protein network captures the evolutionary information, a key limitation is the intensive computation of pairwise sequence similarities needed to construct very large protein networks. In this article, we introduce label propagation on low-rank kernel approximation (LP-LOKA) for searching massively large protein networks. LP-LOKA propagates initial protein similarities in a low-rank graph by Nyström approximation without computing all pairwise similarities. With scalable parallel implementations based on distributed-memory using message-passing interface and Apache-Hadoop/Spark on cloud, LP-LOKA can search protein networks with one million proteins or more. In the experiments on Swiss-Prot/ADDA/CASP data, LP-LOKA significantly improved protein ranking over the widely used HMM-HMM or profile-sequence alignment methods utilizing large protein networks. It was observed that the larger the protein similarity network, the better the performance, especially on relatively small protein superfamilies and folds. The results suggest that computing massively large protein network is necessary to meet the growing need of annotating proteins from newly sequenced species and LP-LOKA is both scalable and accurate for searching massively large protein networks.     

Automated detection of remote homology

The classification of a newly identified protein as a member of a superfamily is important for focusing experiments on its most likely functions. Such classification, often performed by hand, has now been fully automated. This sophisticated new approach takes into account not only alignment scores but also a number of other computable attributes, such as functional sites deduced from sequence conservation patterns.     

STRUCTFAST: protein sequence remote homology detection and alignment using novel dynamic programming and profile-profile scoring

STRUCTFAST is a novel profile-profile alignment algorithm capable of detecting weak similarities between protein sequences. The increased sensitivity and accuracy of the STRUCTFAST method are achieved through several unique features. First, the algorithm utilizes a novel dynamic programming engine capable of incorporating important information from a structural family directly into the alignment process. Second, the algorithm employs a rigorous analytical formula for profile-profile scoring to overcome the limitations of ad hoc scoring functions that require adjustable parameter training. Third, the algorithm employs Convergent Island Statistics (CIS) to compute the statistical significance of alignment scores independently for each pair of sequences. STRUCTFAST routinely produces alignments that meet or exceed the quality obtained by an expert human homology modeler, as evidenced by its performance in the latest CAFASP4 and CASP6 blind prediction benchmark experiments.     

Phylogenetic information improves homology detection.

We present a database search method that is based on phylogenetic trees (treesearch). The method is used to search a protein sequence database for homologs to a protein family. In preparation for the search, a phylogenetic tree is constructed from a given multiple alignment of the family. During the search, each database sequence is temporarily inserted into the tree, thus adding a new edge to the tree. Homology between family and sequence is then judged from the length of this edge. In a comparison of our method to profiles (ISREC pfsearch), two implementations of hidden Markov models (HMMER hmmsearch and SAM hmmscore), and to the family pairwise search (FPS) method on 43 families from the SCOP database based on minimum false-positive counts (min-FPCs), we found a considerable gain in sensitivity. In 69% of the test cases, treesearch showed a min-FPC of at most 50, whereas the two second best methods (hmmsearch and FPS) showed this performance only in 53% cases. A similar impression holds for a large range of min-FPC thresholds. The results demonstrate that phylogenetic information can significantly improve the detection of distant homologies and justify our method as a useful alternative to existing methods.     

A discriminative method for family-based protein remote homology detection that combines inductive logic programming and propositional models

Background  

Remote homology detection is a hard computational problem. Most approaches have trained computational models by using either full protein sequences or multiple sequence alignments (MSA), including all positions. However, when we deal with proteins in the "twilight zone" we can observe that only some segments of sequences (motifs) are conserved. We introduce a novel logical representation that allows us to represent physico-chemical properties of sequences, conserved amino acid positions and conserved physico-chemical positions in the MSA. From this, Inductive Logic Programming (ILP) finds the most frequent patterns (motifs) and uses them to train propositional models, such as decision trees and support vector machines (SVM).     

Efficient remote homology detection using local structure

MOTIVATION: The function of an unknown biological sequence can often be accurately inferred if we are able to map this unknown sequence to its corresponding homologous family. At present, discriminative methods such as SVM-Fisher and SVM-pairwise, which combine support vector machine (SVM) and sequence similarity, are recognized as the most accurate methods, with SVM-pairwise being the most accurate. However, these methods typically encode sequence information into their feature vectors and ignore the structure information. They are also computationally inefficient. Based on these observations, we present an alternative method for SVM-based protein classification. Our proposed method, SVM-I-sites, utilizes structure similarity for remote homology detection. RESULT: We run experiments on the Structural Classification of Proteins 1.53 data set. The results show that SVM-I-sites is more efficient than SVM-pairwise. Further, we find that SVM-I-sites outperforms sequence-based methods such as PSI-BLAST, SAM, and SVM-Fisher while achieving a comparable performance with SVM-pairwise. AVAILABILITY: I-sites server is accessible through the web at http://www.bioinfo.rpi.edu. Programs are available upon request for academics. Licensing agreements are available for commercial interests. The framework of encoding local structure into feature vector is available upon request.     

Protein homology detection and fold inference through multiple alignment entropy profiles

Homology detection and protein structure prediction are central themes in bioinformatics. Establishment of relationship between protein sequences or prediction of their structure by sequence comparison methods finds limitations when there is low sequence similarity. Recent works demonstrate that the use of profiles improves homology detection and protein structure prediction. Profiles can be inferred from protein multiple alignments using different approaches. The "Conservatism-of-Conservatism" is an effective profile analysis method to identify structural features between proteins having the same fold but no detectable sequence similarity. The information obtained from protein multiple alignments varies according to the amino acid classification employed to calculate the profile. In this work, we calculated entropy profiles from PSI-BLAST-derived multiple alignments and used different amino acid classifications summarizing almost 500 different attributes. These entropy profiles were converted into pseudocodes which were compared using the FASTA program with an ad-hoc matrix. We tested the performance of our method to identify relationships between proteins with similar fold using a nonredundant subset of sequences having less than 40% of identity. We then compared our results using Coverage Versus Error per query curves, to those obtained by methods like PSI-BLAST, COMPASS and HHSEARCH. Our method, named HIP (Homology Identification with Profiles) presented higher accuracy detecting relationships between proteins with the same fold. The use of different amino acid classifications reflecting a large number of amino acid attributes, improved the recognition of distantly related folds. We propose the use of pseudocodes representing profile information as a fast and powerful tool for homology detection, fold assignment and analysis of evolutionary information enclosed in protein profiles.     

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.     

Word correlation matrices for protein sequence analysis and remote homology detection

Background  

Classification of protein sequences is a central problem in computational biology. Currently, among computational methods discriminative kernel-based approaches provide the most accurate results. However, kernel-based methods often lack an interpretable model for analysis of discriminative sequence features, and predictions on new sequences usually are computationally expensive.     

RANKPROP: a web server for protein remote homology detection

Summary: We present a large-scale implementation of the RANKPROPprotein homology ranking algorithm in the form of an openlyaccessible web server. We use the NRDB40 PSI-BLAST all-versus-allprotein similarity network of 1.1 million proteins to constructthe graph for the RANKPROP algorithm, whereas previously, resultswere only reported for a database of 108 000 proteins. We alsodescribe two algorithmic improvements to the original algorithm,including propagation from multiple homologs of the query andbetter normalization of ranking scores, that lead to higheraccuracy and to scores with a probabilistic interpretation. Availability: The RANKPROP web server and source code are availableat http://rankprop.gs.washington.edu Contact: iain{at}nec-labs.com; noble{at}gs.washington.edu Associate Editor: Burkhard Rost     

Recent trends in remote homology detection: an Indian Medley

The development of remote homology detection methods is a challenging area in Bioinformatics. Sequence analysis-based approaches that address this problem have employed the use of profiles, templates and Hidden Markov Models (HMMs). These methods often face limitations due to poor sequence similarities and non-uniform sequence dispersion in protein sequence space. Search procedures are often asymmetrical due to over or under-representation of some protein families and outliers often remain undetected. Intermediate sequences that share high similarities with more than one protein can help overcome such problems. Methods such as MulPSSM and Cascade PSI-BLAST that employ intermediate sequences achieve better coverage of members in searches. Others employ peptide modules or conserved patterns of motifs or residues and are effective in overcoming dependencies on high sequence similarity to establish homology by using conserved patterns in searches. We review some of these recent methods developed in India in the recent past.     

Effective detection of remote homologues by searching in sequence dataset of a protein domain fold

Profile matching methods are commonly used in searches in protein sequence databases to detect evolutionary relationships. We describe here a sensitive protocol, which detects remote similarities by searching in a specialized database of sequences belonging to a fold. We have assessed this protocol by exploring the relationships we detect among sequences known to belong to specific folds. We find that searches within sequences adopting a fold are more effective in detecting remote similarities and evolutionary connections than searches in a database of all sequences. We also discuss the implications of using this strategy to link sequence and structure space.     

A novel approach to remote homology detection: jumping alignments.

We describe a new algorithm for protein classification and the detection of remote homologs. The rationale is to exploit both vertical and horizontal information of a multiple alignment in a well-balanced manner. This is in contrast to established methods such as profiles and profile hidden Markov models which focus on vertical information as they model the columns of the alignment independently and to family pairwise search which focuses on horizontal information as it treats given sequences separately. In our setting, we want to select from a given database of "candidate sequences" those proteins that belong to a given superfamily. In order to do so, each candidate sequence is separately tested against a multiple alignment of the known members of the superfamily by means of a new jumping alignment algorithm. This algorithm is an extension of the Smith-Waterman algorithm and computes a local alignment of a single sequence and a multiple alignment. In contrast to traditional methods, however, this alignment is not based on a summary of the individual columns of the multiple alignment. Rather, the candidate sequence is at each position aligned to one sequence of the multiple alignment, called the "reference sequence." In addition, the reference sequence may change within the alignment, while each such jump is penalized. To evaluate the discriminative quality of the jumping alignment algorithm, we compare it to profiles, profile hidden Markov models, and family pairwise search on a subset of the SCOP database of protein domains. The discriminative quality is assessed by median false positive counts (med-FP-counts). For moderate med-FP-counts, the number of successful searches with our method is considerably higher than with the competing methods.     

Adapting genetic regulatory models by genetic programming

In this paper, we focus on the task of adapting genetic regulatory models based on gene expression data from microarrays. Our approach aims at automatic revision of qualitative regulatory models to improve their fit to expression data. We describe a type of regulatory model designed for this purpose, a method for predicting the quality of such models, and a method for adapting the models by means of genetic programming. We also report experimental results highlighting the ability of the methods to infer models on a number of artificial data sets. In closing, we contrast our results with those of alternative methods, after which we give some suggestions for future work.     

A comparison of profile hidden Markov model procedures for remote homology detection

Profile hidden Markov models (HMMs) are amongst the most successful procedures for detecting remote homology between proteins. There are two popular profile HMM programs, HMMER and SAM. Little is known about their performance relative to each other and to the recently improved version of PSI-BLAST. Here we compare the two programs to each other and to non-HMM methods, to determine their relative performance and the features that are important for their success. The quality of the multiple sequence alignments used to build models was the most important factor affecting the overall performance of profile HMMs. The SAM T99 procedure is needed to produce high quality alignments automatically, and the lack of an equivalent component in HMMER makes it less complete as a package. Using the default options and parameters as would be expected of an inexpert user, it was found that from identical alignments SAM consistently produces better models than HMMER and that the relative performance of the model-scoring components varies. On average, HMMER was found to be between one and three times faster than SAM when searching databases larger than 2000 sequences, SAM being faster on smaller ones. Both methods were shown to have effective low complexity and repeat sequence masking using their null models, and the accuracy of their E-values was comparable. It was found that the SAM T99 iterative database search procedure performs better than the most recent version of PSI-BLAST, but that scoring of PSI-BLAST profiles is more than 30 times faster than scoring of SAM models.