Functional characterisation of putative Cis-Regulatory Risk Loci for breast cancer

Rosli, Nordiana

http://hdl.handle.net/10400.1/8544

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
Functional Characterisation of Putative Cis-Regulatory Risk Loci For Breast Cancer by Nordiana Rosli (EMQAL 2013-2015).pdf		2.66 MB	Adobe PDF	Download

Send Feedback

Authors

Rosli, Nordiana

Advisor(s)

Maia, Ana Teresa

Xavier, Joana

Abstract(s)

At present, 94 breast cancer susceptibility loci have been discovered from genome-wide association studies (GWAS). The next step is to identify the causal risk variant, the target gene and to understand the underlying disease mechanism. Studies revealed that most of the variants discovered by GWAS are cis-acting regulatory. Cis-acting regulatory variants can be identified most efficiently by differential allelic expression (DAE) analysis. A DAE genome-wide mapping was done in normal breast tissue, which was cross-compared with GWAS breast cancer data. 19 loci associated with risk and with evidence of cis-regulation were identified, including the 5q14.2 locus that has one SNP associated with risk - rs7707921, and five SNPs displaying DAE across three genes: ATG10, RPS23 and ATP6AP1L. The aim of this thesis is to set out to map the regulatory variants responsible for the DAE signals in the 5q14.2 locus and to determine which one(s) is (are) associated with risk for breast cancer. We performed in silico analysis using data obtained through publically accessible databases, to identify candidate regulatory SNPs (rSNPs) that could be responsible for the DAE and determine if they may be associated with risk to breast cancer. Experimental in vitro analysis by EMSA and analysis of available ChIP-seq data was also conducted in order to investigate possible interactions between candidate rSNPs and transcription factors (TFs). In this study, three SNPs rs226198, rs6880209 and rs17247678 were identified as potential cis-acting regulators of ATG10, RPS23 and ATP6AP1L. Henceforth, we propose a risk model based on our findings: Binding of c-Myc and POL2 to the common allele of rs226198 and rs6880209 lead to over expression of RPS23 and under expression of ATG10, respectively, whereas, binding of STAT3 and c-FOS to rs17247678 lead to under expression of ATP6AP1L, increasing the risk for breast cancer.