Analiza povezanosti somatskih varijanti tumora s mutacijma i mutacijskim potpisima

master's thesis

• Title (english): Association analysis of somatic tumor variants with mutations and mutational signatures
• Author: Ingrid Tomljanović
• Mentor: Fran Supek (mentor)
• Mentor: Dubravko Pavoković (mentor)
• Committee member: Dubravko Pavoković (predsjednik povjerenstva)
• Committee member: Antun Alegro (član povjerenstva)
• Committee member: Marin Ježić (član povjerenstva)
• Committee member: Biljana Balen (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Biology); Zagreb
• Defense date and country: 2019-03-08, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Biology

Abstract

Tijekom života pacijenta, genom stanice raka postupno nakuplja somatske mutacije uslijed
djelovanja raznih endogenih i egzogenih mutacijskih procesa. Genomske otiske ovih procesa
moguće je razlučiti iz podataka o somatskim mutacijama tumora te ih nazivamo mutacijskim
potpisima. Cilj ovog diplomskog rada bio je proučiti povezanost opsežnog skupa somatskih
mutacija koje pripadaju 721 odabranom genu sa 30 mutacijskih potpisa i zamjenskih varijabli
na 33 različita tipa raka. Dvostrano neparametarsko testiranje primijenjeno je na razini svih
tipova raka te pojedinih tipova raka u svrhu testiranja pojedinih ponavljajućih somatskih
mutacija te testiranja tereta mutacija u ponavljajuće mutiranim genima na osnovi podatkovnog
seta TCGA MC3 koji sadrži ~10,000 uzoraka. Analize su podijeljene po statusu
hipermutiranosti uzoraka te predviđenom dobroćudnom ili zloćudnom tipu mutacija. Rezultati
su u svrhu ispravka višestrukog testiranja hipoteza podvrgnuti metodi genomske kontrole te
Benjamini-Hochbergovoj metodi i filtrirani po apsolutnoj vrijednosti veličine učinka. Testiranjem
pojedinih mutacija otkrivene su 62 značajne poveznice, dok su testiranjem tereta mutacija
otkrivene 162 značajne poveznice na razini pojedinih tipova raka i svim podskupovima
podataka. Većina otkrivenih poveznica predstavljaju nova otkrića i moguće vrijedne ciljeve
budućih istraživanja mutacijskih procesa aktivnih u stanicama raka.

Abstract (english)

Throughout the lifetime of a patient, the genome of a cancer cell gradually accumulates
somatic mutations due to a variety of endogenous and exogenous mutational processes. The
genomic fingerprints of those processes can be deconvoluted from tumor somatic mutation
data and are termed mutational signatures. The goal of this master thesis was to assess the
relationships of a comprehensive set of somatic mutations belonging to a list of 721 selected
genes with the 30 mutational signatures and their proxies across 33 different cancer types.
Two-sided nonparametric testing was employed on both pan-cancer and per-cancer type
levels to individually test recurrent somatic mutations and to perform burden testing of
recurrently mutated genes based on the TCGA MC3 dataset comprising ~10,000 samples.
Analyses were stratified by sample hypermutation status and predicted benign or pathogenic
mutation character. Results were submitted to genomic control and Benjamini-Hochberg
multiple testing correction procedures and filtered using absolute effect sizes. Individual
mutation testing identified 62 significant associations, while burden testing identified 162
significant associations on the per-cancer type level across data subsets. The majority of
identified associations were novel findings and constitute potentially valuable targets for future
research into the mutational processes operative in cancer cells.

• Keywords: mutacijski potpisi; TCGA; somatske mutacije; pogonski geni; COSMIC; testiranje tereta; genomska kontrola
• Keywords (english): mutational signatures; TCGA; somatic mutations; driver genes; COSMIC; burden testing; genomic control
• Language: croatian
• URN:NBN: urn:nbn:hr:217:440763
• Study programme: Title: Molecular Biology; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra molekularne biologije (magistar/magistra molekularne biologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access; Embargo expiration date: 2021-03-08
• Created on: 2019-04-10 09:05:17