Molecular profiling to determine aetiology of ascending aortic dilation in bicuspid aortic valve patients

Rationale
Bicuspid aortic valve (BAV) is a congenital abnormality resulting in the fusion of two leaflets of the aortic valve (AV). Whilst this may remain asymptomatic if it continues to function normally, calcification can result in reduced function. The condition also associates with ascending aorta (AA) dilatation (AAD), which may dissect or rupture with fatal consequences. Aortic valve replacement (AVR) surgery for BAV also provides the potential to detect and/or correct TAA, but surgical removal of the dilated AA confers an increased risk. It is important to better understand the mechanisms that confer higher risk for TAA in BAV patients. AVR surgery allows collection of both AV and AA (aneurysmal and non-aneurysmal) samples from the same patients.

This research will integrate within a wider research programme on biomarkers for evolution of AA and targets for therapeutic intervention (developed by Profs Emanueli, Caputo, Angelini and Mr Rajakaruna*).

Aims & Objectives
The objective of this PhD project is to investigate and compare the genetic, epigenetic and molecular profiles between AV and TA in BAV patients undergoing AVR (and presenting TAA of different sizes) and identify BAV and TA molecular changes which may correlate with TAA size.

Aims:
1. Assess genome-wide methylation profiles in tissues
2. Perform genetic association analysis of people undergoing BAV surgery
3. Integrate different types of molecular data

Methods
The project will involve the QC, processing and analysis of genome-wide methylation data (Illumina HM450 array), genotype array data (Illumina HumanCoreExome) and microRNA data from RNA-Seq. The following techniques are likely to be used:

1. Epigenome-wide association analysis (EWAS) to compare methylation profiles between bicuspid valve and thoracic aorta, and to test for association between methylation profiles in the AV and TA tissues and TAA size
2. Genome-wide association analysis (GWAS) to test for association of genetic variants with methylation profiles in valve and aortic tissues. Pathway analysis and other approaches to relate association results to public data on molecular pathways and function
3. Multiple-kernel learning (MKL) and other data integration approaches to combine different types of molecular data to evaluate relative importance of mechanisms
4. Data visualization approaches

* as part of the Bristol Cardiovascular BRU, the Leducq transatlantic network in vascular microRNAs and the BHF Regenerative Medicine Centres