Naslov
Symbolic Cantor algorithm defines secondary protein structure of eukaryotes, prokaryotes and viruses

Autori
Štambuk, Nikola ; Konjevoda, Paško ; Pokrić, Biserka

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
Book of Abstracts MATH/CHEM/COMP/2001 / Graovac, Ante; Pokrić, Biserka; Smrečki, Vilko - Zagreb : Institut Ruđer Bošković, 2001

Skup
MATH/CHEM/COMP/2001 - The 16th Dubrovnik International Course & Conference on the Interfaces among Mathematics, Chemistry and Computer Sciences

Mjesto i datum
Dubrovnik, Hrvatska, 25.06.2001. - 30.06.2001

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
amino acids; Cantor; coding; computational algorithm; eukaryotes; machine learning; nucleotides; prokaryotes; protein folding; structure; viruses

Sažetak
We present new computational algorithm for the secondary protein structure prediction. The method is named Symbolic Cantor Algorithm (SCA), and it is based on the symbolic dynamics and Cantor set coding of the nucleotide and amino acid addresses on the binary tree (Štambuk, N. (2000) Croat. Chem. Acta 73, 1123). The algorithm has two steps. Protein or gene sequence is first transformed into a numerical series of the symbolic coding addresses. The second step of the model involves the extraction of the rules for the recognition of particular folding type by means of the machine learning and spectral analyses of different string parameters. The method was tested on nonhomologous alpha- and beta-protein folds obtained from the Jpred database. It is shown that SCA enables quick, simple and accurate prediction of the secondary protein folding types on a personal computer.

Izvorni jezik
Engleski

Znanstvena područja
Temeljne medicinske znanosti

POVEZANOST RADA