Principal Investigator: Associate Professor Guillaume Lettre
Department: Montreal Heart Institute
Montreal Heart Institute,
Medicine, Research Centre 3rd Floor,
5000 Belanger Street,
Quebec, H1T 1C8, CanadaTags: 11707, blood, genetics, heart diseases, hematology, stroke
1a: The proliferation and differentiation of hematopoietic progenitor cells into mature blood cells is a tightly regulated process. Red blood cell (RBC), white blood cell (WBC) and platelet counts are used in medicine as biomarkers to monitor general health status, to diagnose diseases, and as prognostic indicators of various clinical disorders. The goals of our study are: (1) to identify novel genetic variants associated with blood-cell trait variation in the UK Biobank participants and (2) test if these genetic variants also associate with cardiovascular diseases, including stroke.
1b: Variation in blood-cell traits is observed in various human diseases (e.g. cancer) and is used as predictive marker for heart diseases and stroke. Our project will explore the genetic contribution to inter-individual blood-cell variation, and test whether these genetic factors also influence our risk of heart diseases and stroke.
1c: The number and features of the main blood cells (red blood cells, white blood cells, and platelets) have been measured in all UK Biobank participants. Similarly, the DNA of all UK Biobank participants will be genotyped on the UK Biobank Affymetrix array. We propose to test the correlation between genotypes and inter-individual variation in blood-cell traits using standard genetic association methodologies. When appropriate, we will control blood-cell variables with potential confounders (such as sex, age, cancer status, infectious disease status, kidney or liver disease, etc.).
1d: Full cohort.
In our original application, we proposed to use the UK Biobank dataset for two aims:
1. To identify genetic variation associated with blood-cell traits (routinely measured during complete blood count (CBC)).
2.To further dissect, using genetic and phenotypic data, the link between blood-cell parameters and cardiovascular diseases (CVD).
In Aim #2, we originally proposed to explore how variants associated with bloodcell traits can (alone or within a genetic risk score) help in discriminating between CVD cases and controls (using standard biostatistic methods such as Cox regression). In this aim, we are proposing two changes:
- We plan on extending disease status prediction to more sophisticated methodologies, including machine-learning-based methods (e.g. deep learning).
- Because our interest resides in the predictive value of blood-cell traits (and variants associated with these phenotypes), we would also like to explore how genetic variants (across the genome and not necessarily associated with blood-cell traits) can predict CVD. These analyses would let us know whether predictions are possible. They would also help in describing a baseline model, to which we could then compare performance of models where “weights” are higher for variants associated with blood-cell traits.
These two minor changes do not require additional data and do not impact our original proposal. These analyses will be performed in my lab by members that are already approved by the UK Biobank Project.