Institutional Repository of Vilnius University
Mašininio mokymosi metodų panaudojimas vėžio potipių identifikavimui /
Abstract
dc:descriptionThis bachelor thesis reviews the possibility of subtyping cancer diseases using various machine learning algorithms together with the data of a suitable biomolecular marker. Also, the possibility of establishing a connection between these markers and other cancer characteristics, such as survival and response to therapy, is presented. Glioblastoma Multiforme has been chosen as the research object of this work. It can be subtyped into four subtypes - classical, mezenchymal, neural and proneural -- using micro RNA (miRNA), namely their expression data. Here the theory regarding these subtypes, short history of discovery and genetic profile of glioblastoma are reviewed. In order to assess the ability of subtype classifying, several classifiers and their combinations were tested. Also, an analysis of patient survival and therapy forms has been done, in respect to miRNA expression. For both classifying and clinical data analysis, the clinical and miRNA expression data of 497 patients were used. The final result of this work - a list of 33 miRNAs, which are connected to all of aforementioned features (subtypes, survival, therapy) of glioblastoma.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2018
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Kičiatovas, Dovydas,
- Contributors dc:contributor
-
- Grinis, Irus
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD29420104&prefLang=en_US