Neural networks for full-scale protein sequence classification: Sequence encoding with singular value decomposition
Machine LearningPublished 1 January 1995Open access
Cathy Wu, Michael W. Berry, Sailaja Shivakumar, Jerry McLarty
Citations102
SJR quartileQ1
SJR score1.15
SNIP2.14
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 SVD (singular value decomposition) method, which compresses the long and sparsen-gram input vectors and captures semantics ofn-gram words, has improved the generalization capability of the network.
Abstract
A neural network classification method has been developed as an alternative approach to the search/organization problem of protein sequence databases. The neural networks used are three-layered, feed-forward, back-propagation networks. The protein sequences are encoded into neural input vectors by a hashing method that counts occurrences ofn-gram words. A new SVD (singular value decomposition) method, which compresses the long and sparsen-gram input vectors and captures semantics ofn-gram words, has improved the generalization capability of the network. A full-scale protein classification system has been implemented on a Cray supercomputer to classify unknown sequences into 3311 PIR (Protein Identification Resource) superfamilies/families at a speed of less than 0.05 CPU second per sequence. The sensitivity is close to 90% overall, and approaches 100% for large superfamilies. The system could be used to reduce the database search time and is being used to help organize the PIR protein sequence database.
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.
Journal of the American Society for Information ScienceIndexing by latent semantic analysis
12,677 Citations1990Scott Deerwester, Susan Dumais +3 more
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.
Optimal Brain Damage
2,613 Citations1989Yann LeCun, John S. Denker +1 more
A class of practical and nearly optimal schemes for adapting the size of a neural network by using second-derivative information to make a tradeoff between network complexity and training set error is derived.
Journal of Molecular BiologyPredicting the secondary structure of globular proteins using neural network models
1,234 Citations1988Ning Qian, Terrence J. Sejnowski
It is concluded from computational experiments on real and artificial structures that no method based solely on local information in the protein sequence is likely to produce significantly better results for non-homologous proteins.
Journal of Molecular BiologyImprovements in protein secondary structure prediction by an enhanced neural network
703 Citations1990D G Kneller, Fred E. Cohen +1 more
By adding neural network units that detect periodicities in the input sequence, this work modestly increased the secondary structure prediction accuracy, and the use of tertiary structural class causes a marked increase in accuracy.
Nucleic Acids ResearchUse of the ‘Perceptron’ algorithm to distinguish translational initiation sites in<i>E. coli</i>
669 Citations1982Gary D. Stormo, Thomas D. Schneider +2 more
A "Perceptron" algorithm is used to find a weighting function which distinguishes E. coli translational initiation sites from all other sites in a library of over 78,000 nucleotides of mRNA sequence.
Proceedings of the National Academy of SciencesLocating protein-coding regions in human DNA sequences by a multiple sensor-neural network approach.
658 Citations1991Edward C. Uberbacher, Richard Mural
This work describes a reliable computational approach for locating protein-coding portions of genes in anonymous DNA sequence using a set of sensor algorithms and a neural network to localize the coding regions.
OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information)Parallel distributed processing: explorations in the microstructure of cognition. Volume 1. Foundations
552 Citations1986David E. Rumelhart, James L. McClelland
Proceedings of the National Academy of SciencesProtein secondary structure prediction with a neural network.
512 Citations1989L. Howard Holley, Martin Karplus
A method is presented for protein secondary structure prediction based on a neural network that achieved a maximum overall predictive accuracy of 63% for three states: helix, sheet, and coil.
Nucleic Acids ResearchAutomated assembly of protein blocks for database searching
476 Citations1991Steven Henikoff, Jorja G. Henikoff
The International Journal of Supercomputing ApplicationsLarge-Scale Sparse Singular Value Computations
408 Citations1992Michael W. Berry
Four numerical methods for computing the singular value decomposition (SVD) of large sparse matrices on a multiprocessor architecture are presented and may help advance the development of future out-of-core sparse SVD methods, which can be used to handle extremely large sparsematrices associated with extremely large databases in query-based information-retrieval applications.
Nucleic Acids ResearchThe SWISS-PROT protein sequence data bank, recent developments
150 Citations1993Amos Bairoch, Brigitte Boeckmann
The SWISS-PROT database distinguishes itself from other protein sequence databases by three distinct criteria: protein sequence quality, consistency and utility.
Nucleic Acids ResearchNeural network optimization for<i>E.coli</i>promoter prediction
149 Citations1991Borries Demeler, Guangwen Zhou
Methods for optimizing the prediction of Escherichia coli RNA polymerase promoter sequences by neural networks using the conserved -10 region and -35 region by a neural network are presented.
BiochemistryPrediction of structural and functional features of protein and nucleic acid sequences by artificial neural networks
134 Citations1992Jonathan D. Hirst, Michael J.E. Sternberg
The problems in nucleic acid analysis tackled by neural networks are the prediction of translation initiation sites in Escherichia coli, the recognition of splice junctions in human mRNA, and the Prediction of promoter sites in E. coli.
FEBS LettersA novel approach to prediction of the 3‐dimensional structures of protein backbones by neural networks
134 Citations1990Henrik Bohr, J. Bohr +5 more
Three‐dimensional structures of protein backbones have been predicted using neural networks using backpropagation learning and distance constraints predicted by the trained neural network were utilized to generate a folded conformation of the protein backbone, using a steepest descent minimization approach.
Protein ScienceProtein classification artificial neural system
122 Citations1992Cathy Wu, George Whitson +3 more
The rapid and accurate superfamily classification of unknown proteins based on the information content of the neural interconnections would be valuable to the organization of protein sequence databases and to the gene recognition in large sequencing projects.
Neural NetworksThe optimised internal representation of multilayer classifier networks performs nonlinear discriminant analysis
121 Citations1990Andrew R. Webb, David Lowe
This paper illustrates why a nonlinear adaptive feed-forward layered network with linear output units can perform well as a pattern classification device and shows that minimising the error at the output of the network is equivalent to maximising a particular norm, the network discriminant function, at theoutput of the hidden units.
Journal of Molecular BiologyDetermination of eukaryotic protein coding regions using neural networks and information theory
117 Citations1992Robert Farber, Alan S. Lapedes +1 more
High accuracies stemming from calculations of mutual information (a correlation measure) of spatially separated codons in exons, and in introns are reported, suggesting that dicodon frequencies of adjacent codons are important for intron/exon discrimination.
Methods in enzymology on CD-ROM/Methods in enzymology[6] Searching through sequence databases
92 Citations1990Russell F. Doolittle
This chapter provides a few simple guidelines and hints to sequence searchers about how to make judgments regarding the significance of what has or has not been found when searching through sequence databases.
Nucleic Acids ResearchAn assessment of neural network and statistical approaches for prediction of E.coli Promoter sites
81 Citations1992Paul Horton, Minoru Kanehisa
This work constructed a perceptron type neural network for E. coli promoter prediction and improved its ability to generalize with a new technique for selecting the sequence features shown during training and showed the correlation between the prediction rates of neural networks achieved by previous researchers and the information content of their data sets.
Choice Reviews OnlineComputational molecular biology: sources and methods for sequence analysis
77 Citations1989
This chapter discusses sequence comparison and alignment - the measurement and interpretation of sequence similarity analysis of DNA sequence data prediction of secondary structure of proteins hydrophobicity profiles molecular evolution conclusion - what shall one do with a newly-determined sequence.
Journal of Molecular BiologyA new family of powerful multivariate statistical sequence analysis techniques
72 Citations1991Marin van Heel
The Multivariate Statistical Sequence Analysis principles can be used for a wide spectrum of sequence analysis problems such as: assignment of family memberships to new sequences, validation of new incoming sequences to be entered into the database, prediction of structure from sequence, discrimination of coding from non-coding DNA regions, and automatic generation of an atlas of protein or DNA sequences.
Protein ScienceSelf‐organized neural maps of human protein sequences
72 Citations1994Edgardo A. Ferrán, Bernard Pflugfelder +1 more
A large‐scale application of a method based on artificial neural networks to cluster protein sequences into families to classify the 1,758 human protein sequences stored in the SwissProt database, whose lengths are greater than 50 amino acids is presented.
Nucleic Acids ResearchThe PIR-lnternational databases
67 Citations1993Winona C. Barker, David G. George +3 more
The architecture of the Protein Sequence Database, a number of other PIR-International databases, and mechanisms for providing access to and for distribution of these databases are described.
Nucleic Acids Research<i>Escherichia coli</i>promoters: neural networks develop distinct descriptions in learning to search for promoters of different spacing classes
55 Citations1992Michael C. O’Neill
Back-propagation neural networks were trained to recognize promoter sequences of each of the three major spacing classes found in E. coli with the object of maximizing their ability to generalize while maintaining the level of false positive identifications at a fraction of 1 percent.
Nucleic Acids ResearchBack-propagation and counter-propagation neural networks for phylogenetic classification of ribosomal RNA sequences
46 Citations1994Cathy Wu, Sailaja Shivakumar
A neural network system has been developed for rapid and accurate classification of ribosomal RNA sequences according to phylogenetic relationship and may be extended into a gene identification system for classifying indiscriminately sequenced DNA fragments.
European Journal of BiochemistryComputer analysis of DNA and protein sequences
35 Citations1991Gunnar von Heijne
Some recent trends in the development of theoretical methods for DNA and protein sequence analysis are reviewed, with particular emphasis on the design of new databases, motif searches, sequence alignment algorithms and applications of neural networks.
National Conference on Artificial IntelligenceMega-classification: discovering motifs in massive datastreams
23 Citations1992Nomi L. Harris, Lawrence Hunter +1 more
The development and application of an efficient unsupervised learning procedure for the classification of an unsegmented datastream, given a set of probabilistic binary similarity judgments between regions in the stream, effective on very large databases and tolerates the presence of noise in the similarity judgements.
Computers & ChemistryClassification neural networks for rapid sequence annotation and automated database organization
18 Citations1993Cathy Wu
A neural network classification method has been developed as an alternative approach to the search/organization problem of large molecular databases and could be used as a filter program to reduce the database search time and help organize the molecular sequence databases.