Grant: Doctoral: Vanier Canada Graduate Scholarships
McGill University
$150,000.00
Multi-ethnic polygenic risk scores to improve risk screening of complex diseases for populations of non-European ancestries
EF 2020-21
Résumé
Genetic predisposition to disease is an important risk factor for many complex diseases. Advances in human genetics have enabled the development of efficient genetic scores, named polygenic risk scores, that aggregate information across millions of genetic features to predict disease risks. In combination with effective application of high-dimensional statistics, machine learning and artificial intelligence, the predictive power of recently developed polygenic risk scores is increasing and indicating high clinical utility. However, nearly all of these advances have occurred using data from individuals of European ancestry.
The effectiveness of polygenic risk scores in individuals of non-European ancestries is still largely unknown. But in the cases where it has been evaluated, the predictive performance of polygenic risk scores is substantially attenuated. This is likely due to differences in the underlying genetic basis of disease in different populations, together with varying environmental exposures. To date, there are no clinically-acceptable polygenic risk scores available to people of non-European ancestries, though these populations, including minorities and aboriginals, who support the health care system and its patients, make up very important proportions of the population of Canada and of many other countries.
Therefore, having gathered several large population-based cohorts of different ethnic background, we aim to develop novel machine learning methods, together with open source software packages to improve the accuracy of polygenic risk scores for complex diseases, such as various types of cancer, coronary heart disease, type 2 diabetes, and osteoporosis, for individuals of non-European ancestries. We anticipate that these polygenic risk scores applicable to multiple ethnic groups will help not only realize the goals of generalizability and clinical utility, but also mitigate long-standing health care disparities.
Programme
Doctoral: Vanier Canada Graduate Scholarships
Bénéficiaire
Emplacement
Quebec, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database
Tel que publié
Id
437116
Name
Lu Tianyuan
Region
Québec
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Genetics
Theme2
Social / Cultural / Environmental / Population Health
Prcname2
Vanier Canada Graduate Scholarships CIHR
Projecttitle
Multi-ethnic polygenic risk scores to improve risk screening of complex diseases for populations of non-European ancestries
Programtype2
Studentship Programs
Orgnameinp2
McGill University
Cihramount2
$150,000
Cihrequipment2
$0
Programname2
Doctoral: Vanier Canada Graduate Scholarships
Keyworddelim
Cohort Study; Complex Diseases; Ethnicity; Genetic Risk Prediction; Machine Learning; Population Genetics
Abstract
Genetic predisposition to disease is an important risk factor for many complex diseases. Advances in human genetics have enabled the development of efficient genetic scores, named polygenic risk scores, that aggregate information across millions of genetic features to predict disease risks. In combination with effective application of high-dimensional statistics, machine learning and artificial intelligence, the predictive power of recently developed polygenic risk scores is increasing and indicating high clinical utility. However, nearly all of these advances have occurred using data from individuals of European ancestry.
The effectiveness of polygenic risk scores in individuals of non-European ancestries is still largely unknown. But in the cases where it has been evaluated, the predictive performance of polygenic risk scores is substantially attenuated. This is likely due to differences in the underlying genetic basis of disease in different populations, together with varying environmental exposures. To date, there are no clinically-acceptable polygenic risk scores available to people of non-European ancestries, though these populations, including minorities and aboriginals, who support the health care system and its patients, make up very important proportions of the population of Canada and of many other countries.
Therefore, having gathered several large population-based cohorts of different ethnic background, we aim to develop novel machine learning methods, together with open source software packages to improve the accuracy of polygenic risk scores for complex diseases, such as various types of cancer, coronary heart disease, type 2 diabetes, and osteoporosis, for individuals of non-European ancestries. We anticipate that these polygenic risk scores applicable to multiple ethnic groups will help not only realize the goals of generalizability and clinical utility, but also mitigate long-standing health care disparities.
Competitiondate
202011
Approvedterm2
3 yrs 0 mth
Pinamesdelim
Lu, Tianyuan
Approvedterm
3 ans 0 mois
Cihrcontribution2
~ $150,000~ $0~