Protein Family Classification with Partial Least Squares
Authors:
- 20 December 2006
- journal article
- research article
- Published by American Chemical Society (ACS) in Journal of Proteome Research
- Vol. 6 (2) , 846-853
Abstract
The quality of protein function predictions relies on appropriate training of protein classification methods. Performance of these methods can be affected when only a limited number of protein samples are available, which is often the case in divergent protein families. Whereas profile hidden Markov models and PSI-BLAST presented significant performance decrease in such cases, alignment-free partial least-squares classifiers performed consistently better even when used to identify short fragmented sequences.Keywords
This publication has 24 references indexed in Scilit:
- The Pfam protein families databaseNucleic Acids Research, 2004
- The SWISS-PROT protein knowledgebase and its supplement TrEMBL in 2003Nucleic Acids Research, 2003
- Gapped BLAST and PSI-BLAST: a new generation of protein database search programsNucleic Acids Research, 1997
- CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choiceNucleic Acids Research, 1994
- DNA and peptide sequences and chemical processes multivariately modelled by principal component analysis and partial least-squares projections to latent structuresAnalytica Chimica Acta, 1993
- Suggestions for “safe” residue substitutions in site-directed mutagenesisJournal of Molecular Biology, 1991
- Identification of common molecular subsequencesJournal of Molecular Biology, 1981
- Conformational preferences of amino acids in globular proteinsBiochemistry, 1978
- Amino acid properties and side-chain orientation in proteins: A cross correlation approachJournal of Theoretical Biology, 1975
- The characterization of amino acid sequences in proteins by statistical methodsJournal of Theoretical Biology, 1968