Reference : Insilico genomes for high-throughput sequencing cancer-specific analysis
Dissertations and theses : Doctoral thesis
Life sciences : Genetics & genetic processes
http://hdl.handle.net/10993/22512
Insilico genomes for high-throughput sequencing cancer-specific analysis
English
Killcoyne, Sarah mailto [University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB) > >]
3-Nov-2015
University of Luxembourg, ​Belval, ​​Luxembourg
Docteur en Biologie
145
del Sol Mesa, Antonio mailto
Schneider, Reinhard mailto
Balling, Rudi mailto
Galas, David mailto
Andrade, Miguel mailto
[en] Cancer genomics ; structural variation
[en] As a genomic disease cancer is unique in that the entire genome can be highly unstable, with new mutations accumulating at a rapid rate and massive alterations to the chromosomal structure. Structural aberrations can be highly significant to a patient’s disease, resulting in aberrant proteins that can drive a cancer to progress faster or metastasize. Such aberrations may also have more subtle effects, enabling the cellular population to more rapidly develop drug resistance or simply generate highly diverse populations within a tumor making targeted therapies less effective.
In fact it is these diverse or heterogeneous cellular populations, with highly mutated and frequently structurally aberrant genomes, that make understanding the extent of a tumor genome’s variation so challenging. Large scale sequencing efforts through the Cancer Genome Atlas and the International Cancer Genome Consortium have sequenced thousands of cancer genomes, and while small-scale variants have enabled researchers to begin to trace the evolutionary history and diversity of tumor genomes, large-scale structural variations have continued to be difficult to identify.Current methods and technologies for short-read sequencing generally rely on fitting genomes to a single reference assembly that is assumed to be representative of all individuals. Tumor genomes, which consist of heterogeneous cellular populations with unique aberrations can vary significantly from a ‘normal’ genome. This means that such single references are poor representations of a cancerous cell population, and so methods that rely less directly on the reference offer better opportunities to investigate these aberrations.
In this project, a new method for large-scale structural variant identification, called MultiSieve, is proposed. This method uses prior knowledge to generate and test multiple references for each patient genome. Validation using simulated data establishes the utility of the method, and a comparison with commonly used methods demonstrates that MultiSieve is capable of finding variations often missed by traditional methods and that there are likely to be more structural variants in patients than have been identified previously.
Luxembourg Centre for Systems Biomedicine (LCSB): Computational Biology (Del Sol Group) ; University of Luxembourg: High Performance Computing - ULHPC
Fonds National de la Recherche - FnR
Researchers
http://hdl.handle.net/10993/22512
FnR ; FNR4717849 > Sarah Killcoyne > IGCSA > Insilico genomes for high-throughput sequencing cancer-specific analysis > 01/10/2012 > 14/09/2015 > 2012

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Open access
Thesis_final-postdefense.pdfAuthor postprint40.12 MBView/Open

Bookmark and Share SFX Query

All documents in ORBilu are protected by a user license.