Detekcija varijacija u jednom nukleotidu uporabom metoda dubokog učenja

master's thesis

• Title (english): Detection of Single Nucleotide Variations Using Deep Learning
• Author: Dominik Stanojević
• Mentor: Bojana Dalbelo-Bašić (mentor)
• Committee member: Bojana Dalbelo-Bašić (predsjednik povjerenstva)
• Committee member: Krešimir Križanović (član povjerenstva)
• Committee member: Marko Čupić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2019-07-10, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Detekcija varijacija je veoma važan problem u bioinformatics zbog mnogo razloga. Osim njegove važnosti, problem detekcije varijacija je zanimljiv zbog teške izgradnje efikasnog i preciznog modela. U ovom radu dan je teoretski uvod u problem detekcije varijacija, s posebnim naglaskom na jednonukleotidne varijacije u diploidnim organizmima. U radu su opisani korišteni podaci kao i postupak izgradnje skupa za učenje i testiranje. Također, u radu je opisan prilagođen model ResNet specifično izgrađen za postupak detekcije varijacija. Dani model ostvaruje rezultate usporedive s drugim modelima koristeći manji skup podataka od drugih modela, primjerice modela Clairvoyante.

Abstract (english)

Variant calling, the task of finding variants is extremely important in bioinformatics for various reasons. Expect being important, it is difficult to build efficient and accurate variant caller. This thesis gives theoretical introduction to the variant calling problem, focusing on single nucleotide variants in diploid organisms. The data and dataset generation are described and its application to the custom ResNet model built for variant calling. The model achieves comparable results using only a fraction of data compared to Clairvoyante for predicting single nucleotide variants.

• Keywords: Nukleotidna varijacija; Jednonukleotidna varijacija; Duboko učenje; Bioinformatika; Strojno učenje
• Keywords (english): Nucleotide Variations; Single Nucleotide Variation; SNV; Deep Learning; Bioinformatics; Machine Learning
• Language: english
• URN:NBN: urn:nbn:hr:168:790213
• 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: 2020-01-07 22:07:04