Grant: CIHR Fellowship
McGill University
$135,000.00
Leveraging drug target Mendelian randomization to predict the effect of lipoprotein(a) inhibition
FY 2022-23
Summary
Drugs exert their effects through the product of genes encoded in our DNA. Genetic variation in these genes often have similar effects as the corresponding drugs and represent a rich source of data to develop new drugs or to better understand health impacts of existing drugs. In this project, we hope to use this strategy to better understand a new class of drugs called lipoprotein(a) (pronounced "lp little a") inhibitors. Approximately 20% of the population have, because of their genetics, high levels of lipoprotein(a) in their blood which increases their risk of heart attacks and strokes. A lipoprotein(a) inhibitor is currently being investigated in a clinical trial, but the results will only be available in 2025 and will leave important questions regarding its efficacy in different subgroups unanswered. For example, whether the drug will work better depending on your sex, genetics or use of other medications will not be formally evaluated. To predict the results of this ongoing trial, we will use genetic variants in the gene encoding lipoprotein(a) to mimic the effect of the drug and to study how drug response might change in these important subgroups of individuals.
Because of differences in genealogical ancestry, the genetic makeup of individuals varies and predicting traits like lipoprotein(a) often requires analyzing narrowly defined subgroups of individuals that are genetically similar. Because of this technical limitation of conventional statistical models, many non-white populations are understudied despite data being available as part of large research cohorts that sample individuals in the general population. A second objective of this project is to develop new methods arising from the field of machine learning and artificial intelligence to predict lipoprotein(a) levels in genetically diverse populations. This is important because such a model will be less sensitive to specificities of the majority population and less biased than current approaches.
Department
Program
CIHR Fellowship
Recipient
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database
As published
Id
473838
Name
Legault Marc-Andre
Region
Québec
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Genetics
Theme2
Biomedical
Prcname2
Fellowships - Post-PhD
Projecttitle
Leveraging drug target Mendelian randomization to predict the effect of lipoprotein(a) inhibition
Programtype2
Fellowship Programs
Orgnameinp2
McGill University
Cihramount2
$135,000
Cihrequipment2
$0
Programname2
CIHR Fellowship
Keyworddelim
Cardiovascular Disease; Drug Targets; Genetics; Lipoprotein(A); Machine Learning; Mendelian Randomization
Abstract
Drugs exert their effects through the product of genes encoded in our DNA. Genetic variation in these genes often have similar effects as the corresponding drugs and represent a rich source of data to develop new drugs or to better understand health impacts of existing drugs. In this project, we hope to use this strategy to better understand a new class of drugs called lipoprotein(a) (pronounced "lp little a") inhibitors. Approximately 20% of the population have, because of their genetics, high levels of lipoprotein(a) in their blood which increases their risk of heart attacks and strokes. A lipoprotein(a) inhibitor is currently being investigated in a clinical trial, but the results will only be available in 2025 and will leave important questions regarding its efficacy in different subgroups unanswered. For example, whether the drug will work better depending on your sex, genetics or use of other medications will not be formally evaluated. To predict the results of this ongoing trial, we will use genetic variants in the gene encoding lipoprotein(a) to mimic the effect of the drug and to study how drug response might change in these important subgroups of individuals.
Because of differences in genealogical ancestry, the genetic makeup of individuals varies and predicting traits like lipoprotein(a) often requires analyzing narrowly defined subgroups of individuals that are genetically similar. Because of this technical limitation of conventional statistical models, many non-white populations are understudied despite data being available as part of large research cohorts that sample individuals in the general population. A second objective of this project is to develop new methods arising from the field of machine learning and artificial intelligence to predict lipoprotein(a) levels in genetically diverse populations. This is important because such a model will be less sensitive to specificities of the majority population and less biased than current approaches.
Competitiondate
202210
Approvedterm2
3 yrs 0 mth
Pinamesdelim
Legault, Marc-Andre
Approvedterm
3 ans 0 mois
Cihrcontribution2
~ $135,000~ $0~