Raquel C. de Melo, Carlos Eduardo R. Lopes, Fernando A., Fernandes Jr, Carlos Henrique da Silveira, Marcelo M. Santoro, Rodrigo L. Carceroni,  Wagner Meira Jr, Arnaldo de A. Araújo
Published May 15, 2006
Genet. Mol. Res. 5 (2): 284-308 (2006)

About the authors
Raquel C. de Melo, Carlos Eduardo R. Lopes, Fernando A., Fernandes Jr, Carlos Henrique da Silveira, Marcelo M. Santoro, Rodrigo L. Carceroni,  Wagner Meira Jr, Arnaldo de A. Araújo

Corresponding author
R.C. de Melo
E-mail: raquelcm@dcc.ufmg.br

ABSTRACT

We modeled the problem of identifying how close twoproteins are structurally by measuring the dissimilarity of their  contact maps. These contact maps are colored images, in which the chromatic  information encodes the chemical nature of the contacts. We studied two conceptually distinct image-processing algorithms to measure thedissimilarity between these contact maps; one was a content-based image retrieval method, and the other was based on image registration. In experiments with contact maps constructed from the protein data bank, our approach was able to identify, with greater than 80% precision, instances of monomers of apolipoproteins, globins, plastocyanins, retinol binding proteins and thioredoxins, among the monomers of Protein Data Bank Select. The image registration approach was only slightly more accurate than the content-based image retrieval approach.

Key words: Protein structure, Image-matching, Image registration, Contact maps, Content-based image retrieval