Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Discovery Launch Supplement

Simon Fraser University

$12,500.00

Statistical methods for complex omics data

EF 2024-25

Résumé
Differential abundance (DA) testing is vital to identifying omics markers differentially expressed with the predictors of interest, such as sex, disease status, and environmental effects. Statistical methods play a central role in DA testing in RNA sequencing, one of the most studied omics modalities. Advancements in sequencing technology have enabled the measurements of omes of more modalities, such as genome, proteome, transcriptome, and epigenome, which pose challenges to statistical modeling of these inherently heterogeneous, sparse, and noisy omics data. The long-term goal of my new research program is to develop flexible, efficient, unifying statistical models for bulk and single-cell omics data of all modalities in complex data scenarios. We will develop theoretically justified regression frameworks that model the key characteristics of single-cell and bulk omics data while adjusting for covariate effects and batch variations induced by different donor, cell type, species, and tissue origins. My team will address 3 short-term objectives over the next 5 years: Objective 1: Develop unifying frameworks for heterogenous bulk omics data of all modalities accounting for donor-, species-, tissues-specific variability. We will develop linear mixture model-based (LMM-based) frameworks that can identify DA markers in any type of bulk omics data against categorical and/or continuous predictors of interest, while adjusting for covariates and batch variability through both main and random effects. These frameworks can be further generalized to multi-tissue DA analysis and cross-species DA analysis. Objective 2: Develop statistical models for heterogeneous scRNA-seq data in complex data scenarios. We will develop a zero-inflated Gaussian mixture model (ZIGMM) to capture the key characteristics of scRNA-seq data: (i) excessive number of zero counts, also known as dropouts; and (ii) donor-specific variability through both main and random effects. Objective 3: Develop theoretically justified, generalizable statistical models for sparse, dense, and multi-omics data. We will provide theoretical justifications on the efficiency of ZIGMM score tests for count omics data. Next, we will develop a ZIGMM-based framework that can effectively capture biological signatures in sparse single-cell omics data and generalize it to single-cell multi-omics data. The proposed research provides robust, unifying LMM-based frameworks for DA testing in bulk and single-cell omics data of all modalities in complex data scenarios. By accounting for batch variability through both main and random effects, our proposed methods enable integrative DA analysis of multi-batch, multi-species, multi-cell type, and multi-tissue omics data, which increases the sample size and thus the statistical power. Our research will provide biologists with theoretically and empirically justified statistical models to identify and understand the molecular basis of phenotypic differences.
Programme
Discovery Launch Supplement
Emplacement
British Columbia, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
DGECR-2024-00036
Name nom
Zhang, Lin
Department département
Department of Statistics and Actuarial Science
Organization id
5
Institution établissement
Simon Fraser University
Province en
British Columbia
Province fr
Colombie-Britannique
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
12500
Program id
DGECR
Program name en
Discovery Launch Supplement
Program name fr
Supplément Tremplin vers la découverte
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1508
Committee name en
Mathematics and Statistics
Committee name fr
Mathématiques et 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
3005
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Biostatistics
Research subject fr
Biostatistiques
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Statistical methods for complex omics data
Keywords
statistical genomics; regression; differential expression analysis; differential abundance testing; single-cell omics; heterogeneity; robustness
Application summary
Differential abundance (DA) testing is vital to identifying omics markers differentially expressed with the predictors of interest, such as sex, disease status, and environmental effects. Statistical methods play a central role in DA testing in RNA sequencing, one of the most studied omics modalities. Advancements in sequencing technology have enabled the measurements of omes of more modalities, such as genome, proteome, transcriptome, and epigenome, which pose challenges to statistical modeling of these inherently heterogeneous, sparse, and noisy omics data. The long-term goal of my new research program is to develop flexible, efficient, unifying statistical models for bulk and single-cell omics data of all modalities in complex data scenarios. We will develop theoretically justified regression frameworks that model the key characteristics of single-cell and bulk omics data while adjusting for covariate effects and batch variations induced by different donor, cell type, species, and tissue origins. My team will address 3 short-term objectives over the next 5 years: Objective 1: Develop unifying frameworks for heterogenous bulk omics data of all modalities accounting for donor-, species-, tissues-specific variability. We will develop linear mixture model-based (LMM-based) frameworks that can identify DA markers in any type of bulk omics data against categorical and/or continuous predictors of interest, while adjusting for covariates and batch variability through both main and random effects. These frameworks can be further generalized to multi-tissue DA analysis and cross-species DA analysis. Objective 2: Develop statistical models for heterogeneous scRNA-seq data in complex data scenarios. We will develop a zero-inflated Gaussian mixture model (ZIGMM) to capture the key characteristics of scRNA-seq data: (i) excessive number of zero counts, also known as dropouts; and (ii) donor-specific variability through both main and random effects. Objective 3: Develop theoretically justified, generalizable statistical models for sparse, dense, and multi-omics data. We will provide theoretical justifications on the efficiency of ZIGMM score tests for count omics data. Next, we will develop a ZIGMM-based framework that can effectively capture biological signatures in sparse single-cell omics data and generalize it to single-cell multi-omics data. The proposed research provides robust, unifying LMM-based frameworks for DA testing in bulk and single-cell omics data of all modalities in complex data scenarios. By accounting for batch variability through both main and random effects, our proposed methods enable integrative DA analysis of multi-batch, multi-species, multi-cell type, and multi-tissue omics data, which increases the sample size and thus the statistical power. Our research will provide biologists with theoretically and empirically justified statistical models to identify and understand the molecular basis of phenotypic differences.