Repository logo

Browsing by Author "Ribeiros, Filipa Alexsandra Oleiro Esteves"

Now showing 1 - 1 of 1
  • Unveiling the role of cis-regulatory variation in breast cancer aetiology
    Publication . Ribeiros, Filipa Alexsandra Oleiro Esteves; Maia, Ana Teresa
    In 2020 female breast cancer moved from second to the most commonly diagnosed cancer worldwide. Although an estimated 30% of breast cancer cases are heritable or due to underlying genetic factors, approximately half of the familial risk for breast cancer still remains unknown. Since 2007, continuous efforts from genome-wide association studies (GWAS) and the Collaborative Oncological Gene-Environment Study (COGS) identified low-risk loci that explain up to 18% of the familial relative risk. But, most of the risk-associated variants identified by GWAS are not the true causal variants, and therefore, functional variants and the biological mechanisms underlying breast cancer susceptibility remain largely unknown. Since most of the variants identified by GWAS lie on non-coding genomic regions, risk-associated variants likely have a cis-regulatory function, as shown by several post-GWAS studies focusing on the identification of the causal variants. In this context, the main goal of this project was to develop and use a new and efficient approach to detect target genes and causal variants in known and new breast cancer predisposition loci. Firstly, to address the aforementioned challenges, this work intended to Identify causal variants acting in candidate loci with strong cis-regulatory potential and association with published and unpublished GWAS. Allelic expression (AE) ratios were used as a quantitative variable in case control association studies to understand how genetic variation can control gene expression and to identify cis-regulatory variants and their target genes. For the 17q22 locus two potential regulatory variants – rs17817901 and rs8066588 – altering a miRNA and a transcription factor binding site, respectively, were identified. Additionally, results showed that STXBP4 and COX11 are the most likely target genes in this locus. A significant association was found in normal breast tissue between the preferential expression of the reference alleles of two single nucleotide polymorphisms (SNPs) located on the 17q22 locus – rs17817901 (TOM1L1/COX11) and rs2628315 (STXBP4) –, and increased risk for breast cancer. This association was also observed in blood samples, which shows the possibility of using this approach in the screening of the general population for breast cancer risk. These results showed that integrating AE ratios as a quantitative variable in case-control association studies is a powerful approach to identify novel risk loci. Next, to perform a genome-wide AE analysis from RNA-sequencing (RNA-seq) data, a comprehensive comparison of variant calling pipelines was conducted. Forty-two variant calling pipelines were systematically compared using data from a gold standard and a normal breast tissue sample. This allowed establishing the most suitable analysis pipeline for further studies aiming at precise AE quantification using RNA-seq data. Finally, this work aimed to identify new loci associated with breast cancer risk, using a genome wide approach. RNA-seq data from 12 normal breast tissue samples of healthy women (controls) and 14 breast cancer patients (cases) was analysed and AE ratios were calculated genome-wide across 7,054 genetic variants. Eight candidate variants associated with breast cancer risk were identified, and for those, the previously proposed case-control association analysis using AE ratios was conducted. This identified CDC16 as the strongest candidate new locus associated with breast cancer risk, with a predicted effect size of -1.83 [95% CI=-2.38, -1.14]. Results from this work provide further evidence that cis-regulatory variation plays a major role in breast cancer susceptibility and shows the power of integrating allelic expression data in cancer risk studies, particularly in identifying risk causal variants and their target genes. Furthermore, it presents a novel efficient approach to identify risk – case-control association analysis using AE ratios. Overall, besides providing important new knowledge on the biological mechanism underlying the risk of breast cancer, which will improve the identification and management of the population at risk, it also provides concepts and approaches that are applicable to other cancers and complex diseases
    2021-12-09Doctoral thesis Embargoed access Show more