Vrednovanje modela za predikciju udaljenosti između RNA atoma

master's thesis

• Title (english): Evaluation of RNA Atom Distance Prediction Models
• Author: Jelena Šarić
• Mentor: Mile Šikić (mentor)
• Committee member: Mile Šikić (predsjednik povjerenstva)
• Committee member: Marko Subašić (član povjerenstva)
• Committee member: Krešimir Križanović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-02-17, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

RNA molekule mogu se savijati i uspostavljati veze same sa sobom. S obzirom na to da takve transformacije kao posljedicu daju bogati spektar trodimenzionalnih struktura i funkcija, znanje o takvim strukturama od velike je važnosti.

RNAFold model daje predikciju o trodimenzionalnoj strukturi koristeći metode dubokih neuronskih mreža i prakse koje su pokazale izvanredan napredak u srodnim problemima.

Ovaj diplomski rad nastoji analizirati različite konfiguracije modela kako bi se dobio pregled i smjer u prvim fazama implementacije modela. Promatrajući više metrika performanse, cilj je pronaći povezanosti između arhitekture i performanse modela kako bi se ustanovili faktori koji bi mogli unaprijediti trenutnu implementaciju.

Abstract (english)

RNA molecules can fold and form connections with themselves. Since such transformations lead to various possible three-dimensional structures and a diversity of corresponding functions, the knowledge of such structures is of great importance.

The RNAFold model predicts a three-dimensional RNA structure using deep learning methods and practices that have shown remarkable progress in related problems.

This thesis aims to analyse different model configurations to get an overview and a direction in the first stages of RNAFold’s implementation. By observing multiple performance metrics, the objective is to find a correlation between model architecture and performance, and determine which factors may improve current model implementation.

• Keywords: RNA; savijanje; model; duboko-učenje; analiza
• Keywords (english): RNA; folding; model; deep-learning; analysis
• Language: english
• URN:NBN: urn:nbn:hr:168:349812
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 13:53:56