Title :
Towards identification of thematic overlaps in gene sets
Author :
Banerjee, Nilanjana ; Varadan, Vinay ; Kamalakaran, Sitharthan ; Janevski, Angel
Author_Institution :
Philips Res. North America, Briarcliff Manor, NY, USA
Abstract :
Genomic signatures for disease prognosis are discovered by applying statistical methods on high-throughput data. Enrichment analysis of gene set annotations provide limited understanding of the biological processes. We use curated sets reflecting various biological processes implicated in cancer and the thematic clusters to describe and compare four breast cancer prognostic signatures. We demonstrate an effective and automated use of curated gene sets to identify thematic overlaps in gene sets based on their associated significant Gene Ontology terms. This method enabled us to compare four breast cancer prognostic signatures The differences in the thematic content suggest that the signatures while answering the issue of aggressiveness are capturing different sets of biological processes to achieve efficacy.
Keywords :
cancer; genetics; genomics; medical computing; ontologies (artificial intelligence); patient diagnosis; statistical analysis; tumours; biological process; breast cancer prognostic signature; disease prognosis; enrichment analysis; gene ontology term; gene set annotation; genomic signature; statistical method; thematic overlaps; Assembly; Bioinformatics; Biological processes; Breast cancer; Databases; Diseases; Genomics; North America; Ontologies; Testing;
Conference_Titel :
Genomic Signal Processing and Statistics, 2009. GENSIPS 2009. IEEE International Workshop on
Conference_Location :
Minneapolis, MN
Print_ISBN :
978-1-4244-4761-9
Electronic_ISBN :
978-1-4244-4762-6
DOI :
10.1109/GENSIPS.2009.5174372