Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Discovery Grants Program - Individual

University of Toronto

$14,000.00

Statistical methods for complex genetic data

FY 2010-11

Summary
The overall objective of this research proposal is to develop a catalog of statistical methods and computational tools for genetic studies of complex human disorders. The proposed research focuses on two problems. The primary concern motivating the first research component is the observed difficulty in replicating initial claims of gene discoveries in genetic studies. One main contributing factor is that the traditional estimates of gene-effect size are often grossly upward-biased due to selection of both significant results and competing genetic variants, leading to optimistic power and sample size calculations for the replication studies. This ascertainment bias is also known as the Beavis effect or Winner's curse. The proposed research examine three statistical approaches, resampling-based, likelihood-based and Bayesian methods, designed to reduce the estimation bias without the necessity of an independent replicate sample. The second part of the research proposal concerns joint modeling of multiple correlated variables. In many genetic studies, the trait/disease of interest often cannot be characterized by one single variable of collected subjects; and there are situations in which the outcome of interest cannot be measured directly and a set of surrogate response variables must be used instead. The proposed research is motivated by a Cystic Fibrosis (CF) project that aims to map and isolate modifier genes influencing severity of the CF disease in CF patients. Manifest variables collected include a number of correlated clinical phenotypes. So far, the analyses of genotype/phenotype correlations have been exclusively univariate. The aim of the proposed research is to increase the efficiency of the study by jointly modeling the available correlated phenotypes through the latent variable model (LVM) methodology. The development of the proposed methods and provision of relevant software packages will contribute significantly to the ultimate goal of gene and therapy discovery for complex human disorders.
Program
Discovery Grants Program - Individual
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
250053-2008
Name nom
Sun, Lei
Department département
Dalla Lana School of Public Health
Organization id
31
Institution établissement
University of Toronto
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2010
Competition year année de concours
2008
Award amount
14000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
14
Committee name en
Statistical Sciences
Committee name fr
Statistique
Area of application code
1101
Area of application group en
Health, education and social services
Area of application group fr
Santé publique, éducation et services sociaux
Area of application en
Human health (including medically-related psychological research)
Area of application fr
Santé publique (y compris recherche psychologique liée à la médecine)
Research subject code
3004
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Applied statistics
Research subject fr
Statistique appliquée
Partie
3
Nb partie
5
Application title
Statistical methods for complex genetic data
Keyword
NSERC
Application summary
The overall objective of this research proposal is to develop a catalog of statistical methods and computational tools for genetic studies of complex human disorders. The proposed research focuses on two problems. The primary concern motivating the first research component is the observed difficulty in replicating initial claims of gene discoveries in genetic studies. One main contributing factor is that the traditional estimates of gene-effect size are often grossly upward-biased due to selection of both significant results and competing genetic variants, leading to optimistic power and sample size calculations for the replication studies. This ascertainment bias is also known as the Beavis effect or Winner's curse. The proposed research examine three statistical approaches, resampling-based, likelihood-based and Bayesian methods, designed to reduce the estimation bias without the necessity of an independent replicate sample. The second part of the research proposal concerns joint modeling of multiple correlated variables. In many genetic studies, the trait/disease of interest often cannot be characterized by one single variable of collected subjects; and there are situations in which the outcome of interest cannot be measured directly and a set of surrogate response variables must be used instead. The proposed research is motivated by a Cystic Fibrosis (CF) project that aims to map and isolate modifier genes influencing severity of the CF disease in CF patients. Manifest variables collected include a number of correlated clinical phenotypes. So far, the analyses of genotype/phenotype correlations have been exclusively univariate. The aim of the proposed research is to increase the efficiency of the study by jointly modeling the available correlated phenotypes through the latent variable model (LVM) methodology. The development of the proposed methods and provision of relevant software packages will contribute significantly to the ultimate goal of gene and therapy discovery for complex human disorders.