TY  - THES
T1  - Bioinformatics approaches for cancer research
A1  - Backes,Christina
Y1  - 2010/08/11
N2  - Cancer is the consequence of genetic alterations that influence the behavior of affected cells. While the phenotypic effects of cancer like infinite proliferation are common hallmarks of this complex class of diseases, the connections between the genetic alterations and these effects are not always evident. The growth of information generated by experimental high-throughput techniques makes it possible to combine heterogeneous data from different sources to gain new insights into these complex molecular processes. The demand on computational biology to develop tools and methods to facilitate the evaluation of such data has increased accordingly. To this end, we developed new approaches and bioinformatics tools for the analysis of high-throughput data. Additionally, we integrated these new approaches into our comprehensive C++ framework GeneTrail. GeneTrail presents a powerful package that combines information retrieval, statistical evaluation of gene sets, result presentation, and data exchange. To make GeneTrail';s capabilities available to the research community, we implemented a graphical user interface in PHP and set up a webserver that is world-wide accessible. In this thesis, we discuss newly integrated algorithms and extensions of GeneTrail, as well as some comprehensive studies that have been performed with GeneTrail in the context of cancer research. We applied GeneTrail to analyze properties of tumor-associated antigens to elucidate the mechanisms of antigen candidate selection. Furthermore, we performed an extensive analysis of miRNAs and their putative target pathways and networks in cancer. In the field of differential network analysis, we employed a combination of expression values and topological data to identify patterns of deregulated subnetworks and putative key players for the deregulation. Signatures of deregulated subnetworks may help to predict the sensitivity of tumor subtypes to therapeutic agents and, hence, may be used in the future to guide the selection of optimal agents. Furthermore, the identified putative key players may represent oncogenes, tumor suppressor genes, or other genes that contribute to crucial changes of regulatory and signaling processes in cancer cells and may serve as potential targets for an individualized tumor therapy. With these applications, we demonstrate the usefulness of our GeneTrail package and hope that our work will contribute to a better understanding of cancer. 
KW  - Krebsforschung
KW  - Onkologie
KW  - Tumorantigen
KW  - Bioinformatik
CY  - Saarbrücken
PB  - Universitäts- und Landesbibliothek
AD  - Postfach 151141, 66041 Saarbrücken
UR  - http://scidok.sulb.uni-saarland.de/volltexte/2010/3280
ER  -