Abstract 2893: Integrated genomic meta-analysis of colorectal cancer by elastic-net.

Abstract

Abstract To identify combinatorial sets of known, putative and new cancer drivers responsible for colorectal cancer (CRC) development and other associated specific clinical outcomes, we have developed an integrative analysis method for cancer genome data. This approach is based on the elastic-net algorithm that we have applied to genomic data from the Cancer Genome Atlas (TCGA) Project. Our supervised analysis simultaneously assesses the contribution of i) copy number variation (CNV), ii) gene expression, iii) miRNA, iv) methylation and v) cancer mutations to clinical features. The ongoing TCGA project is generating genomic and clinical data sets from different tumor types including CRC. These detailed catalogues of genetic changes in cancer genomes will continue to provide us with new insights about cancer development. However, extracting biologically/clinically relevant information from TCGA's diverse and large cancer genome data remains a challenge. In attempt to overcome this challenge, we use regularized regression method: elastic-net that improves on “the least absolute shrinkage and selection operator” (Lasso). To demonstrate the performance and validity of this approach, we showed that elastic-net successfully identified i) synthetic genes that have their CNVs perfectly associated with stages from a simulated data ii) TGFBR2 and other driver genes that have mutations associated with CRCs that demonstrate microsatellite instability CRC from TCGA data, and iii) IDH1 that obtained mutations associated with survival according to glioblastoma (GBM) data. In the next phase of the study, we identified the top ranked genes that delineate key clinical features such as TNM stages of CRC. We have identified a series of candidate genes that may indicate clinical stages including novel candidates on chromosome 8. Overall, we have successfully demonstrated that our approach allows for an integrative and highly robust supervised analysis of TCGA data. Citation Format: HoJoon Lee, Patrick Flaherty, Hanlee P. Ji. Integrated genomic meta-analysis of colorectal cancer by elastic-net. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 2893. doi:10.1158/1538-7445.AM2013-2893