Poravnanje proteinskih struktura

master's thesis

• Author: Ivana Kurolt
• Mentor: Pavle Goldstein (mentor)
• Committee member: Pavle Goldstein (predsjednik povjerenstva)
• Committee member: Igor Pažanin (član povjerenstva)
• Committee member: Boris Guljaš (član povjerenstva)
• Committee member: Dijana Ilišević (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Mathematics); Zagreb
• Defense date and country: 2014-09-23, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Mathematics

Abstract

U ovom radu opisan je jedan algoritam za poravnanje proteinske strukture različitih duljina. Na početku smo definirali matrice udaljenosti točaka za dvije proteinske strukture zasebno. Pomoću tih matrica smo napravili novu koja uspoređuje njihove vektore određene duljine (koju smo sami zadali parametrom wind) i govori koliko su bliski. Kako bismo odredili što je relativno blisko, uveli smo novi parametar (trash), te sve vrijednosti koje su manje od parametra postavili na 1, a ostale na 0. Nakon toga smo tražili uzastopne pozicije, koje u matrici imaju vrijednost 1, i koje zapravo predstavljaju točke iz dva skupa točaka sa početka. Pozicije najduljih nizova smo spremili u polje iz kojeg smo kasnije izbacivali one koji su najlošiji prema ostalima. Tim postupkom smo došli do pozicija u proteinima koje međusobno imaju najmanje udaljenosti iz kojih zaključujemo na kojim dijelovima se proteini podudaraju, odnosno na kojim dijelovima su poravnati.

Abstract (english)

This paper describes an algorithm for protein structure alignment of different lengths. At the beginning, we define the matrix of distances between points for two protein structures separately. Using these matrices, we made a new one that compares their vectors of a certain length (which we have set with parameter wind) and tells us how close they are. To determine which is relatively close, we introduced a new parameter (trash), and all values that are less than the parameter set to 1, and the rest to 0. After that we were looking for consecutive positions, which in the matrix have a value 1, and that actually represent points from the two sets of points from the beginning. Positions of longest strings are stored in the field, from which we later evicted those who are worst to others. With this process, we come to positions in proteins that have the least distance from each other. From that, we conclude what parts of the proteins coincide, ie which parts are aligned.

• Keywords: poravnavanje proteinskih struktura; matrice udaljenosti točaka
• Keywords (english): protein structure alignment; matrix of distances between points
• Language: croatian
• URN:NBN: urn:nbn:hr:217:806932
• Study programme: Title: Applied Mathematics; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra matematike (magistar/magistra matematike)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Repository: Repository of the Faculty of Science
• Created on: 2019-02-21 12:58:21