A Discriminative Framework for Detecting Remote Protein Homologies
Journal of Computational BiologyPublished 1 February 2000
Tommi Jaakkola, Mark Diekhans, David Haussler
Citations494
SJR quartileQ2
SJR score0.63
SNIP0.52
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
A new method for detecting remote protein homologies is introduced and shown to perform well in classifying protein domains by SCOP superfamily using a new kernel function derived from a generative statistical model for a protein family, in this case a hidden Markov model.
Abstract
A new method for detecting remote protein homologies is introduced and shown to perform well in classifying protein domains by SCOP superfamily. The method is a variant of support vector machines using a new kernel function. The kernel function is derived from a generative statistical model for a protein family, in this case a hidden Markov model. This general approach of combining generative models like HMMs with discriminative methods such as support vector machines may have applications in other areas of biosequence analysis as well.
Keywords
Biochemistry, Genetics and Molecular Biology
Journal of Molecular BiologyBasic local alignment search tool
94,539 Citations1990Stephen F. Altschul, Warren Gish +3 more
A new approach to rapid sequence comparison, basic local alignment search tool (BLAST), directly approximates alignments that optimize a measure of local similarity, the maximal segment pair (MSP) score.
Nucleic Acids ResearchGapped BLAST and PSI-BLAST: a new generation of protein database search programs
74,499 Citations1997Stephen F. Altschul
A new criterion for triggering the extension of word hits, combined with a new heuristic for generating gapped alignments, yields a gapped BLAST program that runs at approximately three times the speed of the original.
The Nature of Statistical Learning Theory
39,279 Citations1995Vladimir Vapnik
Proceedings of the National Academy of SciencesImproved tools for biological sequence comparison.
11,334 Citations1988William R. Pearson, David J. Lipman
Three computer programs for comparisons of protein and DNA sequences can be used to search sequence data bases, evaluate similarity scores, and identify periodic structures based on local sequence similarity.
International Conference on Neural Information ProcessingAdvances in kernel methods: support vector learning
5,815 Citations1999Bernhard Schölkopf, Christopher J. C. Burges +1 more
Support vector machines for dynamic reconstruction of a chaotic system, Klaus-Robert Muller et al pairwise classification and support vector machines, Ulrich Kressel.
IEEE ASSP MagazineAn introduction to hidden Markov models
4,789 Citations1986L. R. Rabiner, Biing‐Hwang Juang
The purpose of this tutorial paper is to give an introduction to the theory of Markov models, and to illustrate how they have been applied to problems in speech recognition.
Biological sequence analysis probabilistic models of proteins and nucleic acids
3,654 Citations2010Richard Durbin
Journal of Molecular BiologyHidden Markov Models in Computational Biology
1,961 Citations1994Anders Krogh, Michael Brown +3 more
The results suggest the presence of an EF-hand calcium binding motif in a highly conserved and evolutionary preserved putative intracellular region of 155 residues in the alpha-1 subunit of L-type calcium channels which play an important role in excitation-contraction coupling.
Nature GeneticsIdentification of protein coding regions by database similarity search
1,688 Citations1993Warren Gish, David J. States
The computer program BLASTX performed conceptual translation of a nucleotide query sequence followed by a protein database search in one programmatic step and was characterized as appropriate for use in moderate and large scale sequencing projects at the earliest opportunity, when the data are most prone to containing errors.
Proceedings of the National Academy of SciencesProfile analysis: detection of distantly related proteins.
1,339 Citations1987Michael Gribskov, A. McLachlan +1 more
Tests with globin and immunoglobulin sequences show that profile analysis can distinguish all members of these families from all other sequences in a database containing 3800 protein sequences.
Exploiting Generative Models in Discriminative Classifiers
1,285 Citations1998Tommi Jaakkola, David Haussler
A natural way of achieving this combination by deriving kernel functions for use in discriminative methods such as support vector machines from generative probability models is developed.
BioinformaticsHidden Markov models for detecting remote protein homologies.
1,111 Citations1998Kevin Karplus, Curtis L. Barrett +1 more
A new hidden Markov model method (SAM-T98) for finding remote homologs of protein sequences is described and evaluated, which is optimized to recognize superfamilies, and would require parameter adjustment to be used to find family or fold relationships.
Nucleic Acids ResearchSCOP: a Structural Classification of Proteins database
960 Citations1997Tim Hubbard, Alexey G. Murzin +2 more
The Structural Classification of Proteins (SCOP) database provides a detailed and comprehensive description of the relationships of known protein structures that provide the basis of the ASTRAL sequence libraries that can be used as a source of data to calibrate sequence search algorithms and for the generation of statistics on, or selections of, protein structures.
Methods in enzymology on CD-ROM/Methods in enzymology[27] Local alignment statistics
724 Citations1996Stephen F. Altschul, Warren Gish
Proceedings of the National Academy of SciencesPrediction of protein folding class using global description of amino acid sequence.
650 Citations1995Inna Dubchak, Ilya Muchnik +2 more
A method for predicting protein folding class based on global protein chain description and a voting process, achieved by a computer-simulated neural network trained on a data base consisting of 83 folding classes, shows that proteins were assigned to the correct class with an average accuracy.
Journal of Molecular BiologySequence comparisons using multiple sequences detect three times as many remote homologues as pairwise methods
563 Citations1998Jong Moon Park, Kevin Karplus +5 more
The extent to which the SAM-T98 implementation of a hidden Markov model procedure; PSI-BLAST; and the intermediate sequence search (ISS) procedure can detect evolutionary relationships between the members of the sequence database PDBD40-J is determined.
Proceedings of the National Academy of SciencesHidden Markov models of biological primary sequence information.
458 Citations1994Pierre Baldi, Yves Chauvin +2 more
A smooth and convergent algorithm is introduced to iteratively adapt the transition and emission parameters of the models from the examples in a given family, yielding an effective multiple-alignment algorithm which requires O(KN2) operations, linear in the number of sequences.
Computer applications in the biosciencesHidden Markov models for sequence analysis: extension and analysis of the basic method
442 Citations1996Richard Hughey, Anders Krogh
Journal of Molecular BiologyPosition-based sequence weights
426 Citations1994Steven Henikoff, Jorja G. Henikoff
This work describes a different approach, which bases weights on the diversity observed at each position in the alignment, rather than on a sequence distance measure, which makes minimal assumptions, is simple to compute, and performs well in comprehensive evaluations.
Journal of Molecular BiologyIdentification of protein sequence homology by consensus template alignment
282 Citations1986William R. Taylor
A pattern-matching procedure is described, based on fitting templates to the sequence, which allows general structural constraints to be imposed on the patterns identified, which demonstrated the specificity of the templates was demonstrated by their ability to identify the conserved features in known immunoglobulin and immunoglOBulin-related sequences but not in other non-immunoglobulus sequences.
Journal of Computational BiologyMaximum Discrimination Hidden Markov Models of Sequence Consensus
246 Citations1995Sean R. Eddy, Graeme Mitchison +1 more
The maximum discrimination method for building hidden Markov models (HMMs) of protein or nucleic acid primary sequence consensus compensates for biased representation in sequence data sets, superseding the need for sequence weighting methods.
Nucleic Acids ResearchExtracting protein alignment models from the sequence database
237 Citations1997Andrew F. Neuwald, Jun S. Liu +2 more
Journal of Molecular BiologyHydration of Proteins
232 Citations1993Roger M. Brunne, Edgars Liepinsh +3 more
Protein ScienceComprehensive assessment of automatic structural alignment against a manual standard, the scop classification of proteins
223 Citations1998Mark Gerstein, Michael Levitt
A simple method for aligning protein sequences on the basis of a 3D structure, on a large scale, to the proteins in the scop classification of fold families is applied, with good agreement and detailed comparison highlights how particular protein structural features are problematical to align.
Journal of Computational BiologyMethods and Statistics for Combining Motif Match Scores
202 Citations1998Timothy L. Bailey, Michael Gribskov
It is shown that combining motif scores indeed gives better search accuracy, and that the MAST sequence homology search algorithm utilizing the product of p-values scoring method is available for interactive use and downloading.
Proteins Structure Function and BioinformaticsPredicting protein structure using hidden Markov models
153 Citations1997Kevin Karplus, Kimmen Sjölander +6 more
How methods based on hidden Markov models performed in the fold‐recognition section of the CASP2 experiment is discussed, with a good score on both methods indicating a high probability that the target sequence is homologous to the structure.
Neural ComputationA New Discriminative Kernel from Probabilistic Models
129 Citations2002Koji Tsuda, Motoaki Kawanabe +3 more
This work proposes a new discriminative TOP kernel derived from tangent vectors of posterior log-odds and develops a theoretical framework on feature extractors from probabilistic models and uses it for analyzing the TOP kernel.
SAM: SEQUENCE ALIGNMENT AND MODELING SOFTWARE SYSTEM
89 Citations1995Richard Hughey, Anders Krogh
The Sequence Alignment and Modeling system (SAM) is a collection of flexible software tools for creating, refining, and using linear hidden Markov models for biological sequence analysis.
Computer applications in the biosciencesScoring hidden Markov models
88 Citations1997Christian Barrett, Richard Hughey +1 more
Among the null model choices, a simple looping null model that emits characters according to the geometric mean of the character probabilities in the columns modeled by the hidden Markov model (HMM) performs well or best across all four discrimination experiments.
Journal of Molecular BiologyGlobal self-organization of all known protein sequences reveals inherent biological signatures 1 1Edited by F. E. Cohen
60 Citations1997Michal Linial, Nathan Linial +2 more
A global classification of all currently known protein sequences is performed, and a novel self-organized, cross-validated clustering algorithm leads to a novel representation of protein families, from which functional biological kinship ofprotein families can be deduced, as demonstrated for the transporter family.
Journal of Computational BiologyProtein Family Classification Using Sparse Markov Transducers
55 Citations2003Eleazar Eskin, William Stafford Noble +1 more
Efficient data structures are described and used to improve the memory usage of SMTs, which outperform previous probabilistic suffix tree methods and under certain conditions perform comparably to state-of-the-art protein homology methods.
PubMedProtein folding class predictor for SCOP: approach based on global descriptors.
55 Citations1997Inna Dubchak, Ilya Muchnik +1 more
This work demonstrates new techniques developed for the prediction of protein folding class in the context of the most comprehensive Structural Classification of Proteins (SCOP) in terms of the physical, chemical and structural properties of its constituent amino acids.
BioinformaticsWeighting hidden Markov models for maximum discrimination.
40 Citations1998Rachel Karchin, Richard Hughey
This new version of the SAM HMM suite outperforms HMMer (17% reduction over PSW for 10 training sequences), Meta-MEME (28% reduction), and unweighted SAM (31% reduction) and this work evaluates sequence weighting methods based on the maximum-discrimination idea.
Proteins Structure Function and BioinformaticsBlind predictions of local protein structure in CASP2 targets using the I‐sites library
34 Citations1997Christopher Bystroff, David Baker
Journal of Computational BiologyA Learning Method of Hidden Markov Models for Sequence Discrimination
25 Citations1996Hiroshi Mamitsuka
It is concluded that both the use of negative examples and the method of using negative examples are useful for training HMMs in discriminating unknown sequences.
Family-based homology detection via pairwise sequence comparison
22 Citations1998William Noble Grundy
A linear combination of multiple pairwise sequence comparisons outperforms motif-based and HMM methods for all query set sizes, and heuristic pairwise comparison algorithms are much more heuristic than the training algorithms for statistical models.
Current Opinion in Structural BiologyAdvances in sequence analysis
13 Citations2001Andrea Califano
The integration of sequence data with a variety of alternative information is providing valuable and fundamentally new insight into biological processes, as well as an array of new computational methodologies for the analysis of biological data.