Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Discovery Grants Program - Individual

Queen's University

$12,000.00

Computational methods in the statistical analysis of microarray data and survival data

EF 2009-10

Résumé
My proposed research program centers on three topics. First, I am interested in developing computer-intensive methods to analyze microarray gene expression data. A microarray study typically monitors the expression levels of tens of thousands of genes on a relatively small number of samples. When the data were used to predict for patient classes with respect to a clinical outcome, it was customary to report only a point estimate for prediction error as a measure of prediction accuracy. I will develop new methods to construct confidence intervals for the prediction error, with applications to microarray cancer classification. A confidence interval for prediction error allows health investigators to assess the variability of the prediction accuracy, which is particularly meaningful for small samples. Second, I am interested in improving the design and analysis of randomized clinical trials when a treatment only benefits a subset of patients (sensitive subset). Treatment effect of this type may be overlooked in traditional clinical trials. Suppose the sensitive subset is determined by unknown gene expression features. In a new design, I will detect treatment effects either through a test on the overall patients (and stop if effective), or through a test on the sensitive subset identified through the gene expression data. The new approach aims to reduce the false rejection rate of clinical trials for testing molecularly targeted treatment. Finally, I plan to develop resampling methods for the analysis of failure time data with measurement error problems. I will apply the new methods to a practical dataset arising from a breast cancer study comparing the effect of two chemotherapy treatments on patients' survival time. The study measured a genetic factor of the patients with possible errors. The variance estimator is either not derived or has a rather complex expression in the existing analysis methods for such problems. Resampling methods will be practically appealing under these circumstances to avoid involved mathematical derivations.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
356016-2008
Name nom
Jiang, Wenyu
Department département
Mathematics and Statistics
Organization id
29
Institution établissement
Queen's University
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2009
Competition year année de concours
2008
Award amount
12000
Program id
RGPIN
Program nane 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
1200
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Advancement of knowledge
Area of application fr
Avancement des connaissances
Research subject code
3005
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Biostatistics
Research subject fr
Biostatistiques
Partie
2
Nb partie
5
Application title
Computational methods in the statistical analysis of microarray data and survival data
Keyword
NA
Application summary
My proposed research program centers on three topics. First, I am interested in developing computer-intensive methods to analyze microarray gene expression data. A microarray study typically monitors the expression levels of tens of thousands of genes on a relatively small number of samples. When the data were used to predict for patient classes with respect to a clinical outcome, it was customary to report only a point estimate for prediction error as a measure of prediction accuracy. I will develop new methods to construct confidence intervals for the prediction error, with applications to microarray cancer classification. A confidence interval for prediction error allows health investigators to assess the variability of the prediction accuracy, which is particularly meaningful for small samples. Second, I am interested in improving the design and analysis of randomized clinical trials when a treatment only benefits a subset of patients (sensitive subset). Treatment effect of this type may be overlooked in traditional clinical trials. Suppose the sensitive subset is determined by unknown gene expression features. In a new design, I will detect treatment effects either through a test on the overall patients (and stop if effective), or through a test on the sensitive subset identified through the gene expression data. The new approach aims to reduce the false rejection rate of clinical trials for testing molecularly targeted treatment. Finally, I plan to develop resampling methods for the analysis of failure time data with measurement error problems. I will apply the new methods to a practical dataset arising from a breast cancer study comparing the effect of two chemotherapy treatments on patients' survival time. The study measured a genetic factor of the patients with possible errors. The variance estimator is either not derived or has a rather complex expression in the existing analysis methods for such problems. Resampling methods will be practically appealing under these circumstances to avoid involved mathematical derivations.