Grant: Discovery Grants Program - Individual
The University of Western Ontario
$27,000.00
Evaluation and Development of High-throughput Covariate Summaries for the Design and Analysis of Two-phase Next-generation Sequencing Studies
EF 2024-25
Résumé
Introduction
The two-phase study uses data collected in phase 1 (e.g. outcomes, auxiliary covariates) to strategically select an informative subsample. In phase 2, expensive information, e.g. next-generation sequencing (NGS) data, is collected in the subsample only reducing the overall cost. Lastly, available data from phases 1 and 2 are jointly analyzed to make inference on parameters modelling the expensive data. This strategy provides a desirable compromise by economically using limited resources such as budget and sample availability while maintaining high statistical performance. Despite the appeal for two-phase NGS studies, there is limited understanding on how to leverage inexpensive high-throughput auxiliary covariates, e.g. genotyping arrays, from previously conducted genome-wide association studies (GWAS) to inform phase 2 subsample selection. In addition, different summaries of (meta)genomic information have been proposed in the literature but their role in designing and analyzing two-phase NGS studies is yet to be studied.
Objective and Aims
The objective of the proposed research is to identify and evaluate approaches that summarize high-throughput auxiliary covariates to inform phase 2 subsampling and to develop novel statistical, machine learning, and computational methodologies that incorporate such summaries for the design and analysis of two-phase NGS studies. The specific aims of the proposal are:
1. To assess dimension reduction and variable selection techniques for post-GWAS approaches, e.g. polygenic risk scores and predicted gene expression, that inform phase 2 subsample selection for inference on NGS data for univariate outcomes.
2. To extend the approaches in 1. for multivariate outcomes of equal interest or univariate outcomes collected longitudinally.
3. To evaluate community summaries of inexpensive microbiome data (16S rRNA sequencing) to inform subsampling of shotgun metagenomics sequencing (SMS) for microbial association studies, impute missing-by-design SMS data via generative adversarial networks, and test for SMS-outcome association through kernel-machine methods.
Novelty and expected significance
There is no consensus on how to best summarize the wide availability of high-throughput auxiliary covariates from phase 1 to identify informative phase 2 subsamples in two-phase NGS studies. Thus, there is an outstanding need for novel methodologies leveraging high-dimensional phase 1 data to measure expensive NGS data in the most informative subsamples only, reducing the overall cost of these studies.
Ultimately, these developments will converge under a multiomic data integration framework for multi-phase designs in which different omics data are collected across partially overlapping informative subsamples, providing a paradigm shift that promotes and champions an economical and strategical use of resources to be targeted for the more expensive NGS data acquisition without sacrificing statistical performance.
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
RGPIN-2023-03602
Name nom
Espin Garcia, Osvaldo
Department département
Epidemiology and Biostatistics
Organization id
36
Institution établissement
The University of Western Ontario
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2023
Award amount
27000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
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
1211
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Medical and health sciences
Area of application fr
Sciences médicales et de la santé
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
Evaluation and Development of High-throughput Covariate Summaries for the Design and Analysis of Two-phase Next-generation Sequencing Studies
Keywords
two-phase studies; next-generation sequencing; outcome-dependent sampling; covariate-dependent sampling; high-throughput auxiliary covariates; post-genome wide association studies; polygenic risk scores; predicted gene expression; multivariate outcomes; microbiome data
Application summary
Introduction
The two-phase study uses data collected in phase 1 (e.g. outcomes, auxiliary covariates) to strategically select an informative subsample. In phase 2, expensive information, e.g. next-generation sequencing (NGS) data, is collected in the subsample only reducing the overall cost. Lastly, available data from phases 1 and 2 are jointly analyzed to make inference on parameters modelling the expensive data. This strategy provides a desirable compromise by economically using limited resources such as budget and sample availability while maintaining high statistical performance. Despite the appeal for two-phase NGS studies, there is limited understanding on how to leverage inexpensive high-throughput auxiliary covariates, e.g. genotyping arrays, from previously conducted genome-wide association studies (GWAS) to inform phase 2 subsample selection. In addition, different summaries of (meta)genomic information have been proposed in the literature but their role in designing and analyzing two-phase NGS studies is yet to be studied.
Objective and Aims
The objective of the proposed research is to identify and evaluate approaches that summarize high-throughput auxiliary covariates to inform phase 2 subsampling and to develop novel statistical, machine learning, and computational methodologies that incorporate such summaries for the design and analysis of two-phase NGS studies. The specific aims of the proposal are:
1. To assess dimension reduction and variable selection techniques for post-GWAS approaches, e.g. polygenic risk scores and predicted gene expression, that inform phase 2 subsample selection for inference on NGS data for univariate outcomes.
2. To extend the approaches in 1. for multivariate outcomes of equal interest or univariate outcomes collected longitudinally.
3. To evaluate community summaries of inexpensive microbiome data (16S rRNA sequencing) to inform subsampling of shotgun metagenomics sequencing (SMS) for microbial association studies, impute missing-by-design SMS data via generative adversarial networks, and test for SMS-outcome association through kernel-machine methods.
Novelty and expected significance
There is no consensus on how to best summarize the wide availability of high-throughput auxiliary covariates from phase 1 to identify informative phase 2 subsamples in two-phase NGS studies. Thus, there is an outstanding need for novel methodologies leveraging high-dimensional phase 1 data to measure expensive NGS data in the most informative subsamples only, reducing the overall cost of these studies.
Ultimately, these developments will converge under a multiomic data integration framework for multi-phase designs in which different omics data are collected across partially overlapping informative subsamples, providing a paradigm shift that promotes and champions an economical and strategical use of resources to be targeted for the more expensive NGS data acquisition without sacrificing statistical performance.