Using integrative omics to disentangle causal relationships between tissue-specific pathways and coronary artery disease

Background

One of the aims of personalised medicine to stratify patients with the same disease into different groups based on their genotypes in order improve treatment choices and patient management. One of the largest limitations to this is that although many genetic variants associated to many complex phenotypes have been identified, how they relate to each other and in which tissue they carry out their function is still in many cases unknown. Knowing which pathways in which tissue affect a disease would help to identify new drugs but would also allow us to understand the differences between each patient which apparently have the exact same disease, making personalised medicine much closer than today. A possible approach to achieve this goal is to combine the genetic variants which underlie the expression of genes in a specific pathway in a specific tissue to predict a target close to its physiological function (i.e. a measured protein) and use Mendelian randomisation to verify the causal effect of the pathway on common disorders such as CAD and the metabolic syndrome.

Aims

The proposed project will thus be focused on creating predictive genetic scores for the overall functionality of biological pathways in specific tissues and in understanding their effect on CAD, as well as traits related to the metabolic syndrome.

Specific aims:

1) Estimation of tissue-specific expression genetic scores for each gene.

The first step of the project will be to create the best SNP-based prediction score using tissue-specific eQTLs in order to maximise the variance explained by the score. To this end, cis-eQTL data from different sources (e.g. Gtex database and larger datasets available via collaborators) will be integrated and re-analysed.

2) Definition of known pathways and tissue-specific ones.

The term “pathway” is used here in the broadest sense, to describe a group of genes, proteins and other mediators that share some meaningful biological similarity. In our view, more granular definitions require reductionist dissection of specific interactions and responses to perturbation. Our approach will combine information from two conceptual sources of pathway annotations: curated pathways, and data-driven inferences. We will use curated pathways from KEGG, Reactome, Wikipathways and other publicly available sources. Moreover, we will also use gene-gene expression relationships from cell-type specific transcriptome atlases, including FANTOM5 and GTEx, to infer meaningful groupings of genes.

3) Estimating the weight of each gene expression score on the overall pathway functionality.

To estimate the overall functionality of a pathway we will use a multivariate model where a known end point for a specific pathway (e.g. IL6 for the Toll Like receptor pathway) will be used to weight the relative effect of each gene expression score on the overall pathway functionality. This procedure will be performed in a tissue-specific manner using only cis-eQTL scores from a specific tissue. Two separated datasets will be used for training and testing the model.

4) Mendelian Randomisation against CAD and Metabolic Syndrome-related traits.

The polygenic risk scores created as outcomes of the previous steps will be used as MR instruments to understand the causal relationship between specific pathways in tissues and the target disease. Given the strong biological knowledge behind the scores, the identified pathway/disease relationships will be causal and thus will give great insight in disease aetiology. Finally, for each patient it will be possible to construct pathway tissue scores, thus stratifying them and potentially helping in defining personalised treatments.

Training Outcomes

Genetic association studies
Predictive model estimation and validation
Bioinformatics analysis of biological pathways through integrating different omic data
Causal inference through Mendelian Randomisation

This MRC programme is joint between the Universities of Edinburgh and Glasgow. You will be registered at the host institution of the primary supervisor detailed in your project selection.

All applications should be made via the University of Edinburgh, irrespective of project location:

http://www.ed.ac.uk/studying/postgraduate/degrees/index.php?r=site/view&id=919

Please note, you must apply to one of the projects and you are encouraged to contact the primary supervisor prior to making your application. Additional information on the application process if available from the link above.

For more information about Precision Medicine visit:

http://www.ed.ac.uk/usher/precision-medicine