Peptide identification using peptide amino acid attribute vectors

J Proteome Res. 2004 Jul-Aug;3(4):813-20. doi: 10.1021/pr0499444.


We describe the theoretical basis for a peptide identification method wherein peptides are represented as vectors based on their amino acid composition and grouped into clusters. Unknown peptides are identified by finding the database cluster and peptide entries with the shortest Euclidian distance. We demonstrate that the amino acid composition of peptides is virtually as informative as the sequence and allows rapid peptide identification more accurately than peptide mass alone.

Publication types

  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Algorithms
  • Animals
  • Computational Biology / methods
  • Humans
  • Peptides / analysis*
  • Peptides / chemistry
  • Proteins / chemistry
  • Proteomics / methods*
  • Rats
  • Sequence Analysis, Protein


  • Peptides
  • Proteins