Grant: Discovery Grants Program - Individual
University of Guelph
$34,000.00
Statistical and computational methods for environment, life, and health.
EF 2024-25
Résumé
Statistical and computational tools are indispensable for supporting research and tackling real world challenges in the fields of environment, life, and health sciences. My long-term goal is to develop and advance statistical and computational tools to address issues and analyze data arising from cutting-edge research in these disciplines. My research program has two main themes: 1) statistical methodology development for microbiome data analysis, and 2) environmental and climate data modelling.
Theme 1: The rapid growth of microbial metagenomics is attributable to the emergence of next generation sequencing. Massive amount of metagenomic sequencing data are available for exploring and disentangling the relationships between biological/environmental factors and gut microbiome compositions, and their impacts on host health. The proposed research program provides novel statistical methodologies in regularized regression models that overcome some of the known challenges existing in the field and provide effective and practical statistical tools for analyzing microbiome composition data. The proposed methods based on the formulation and utility of minorization-maximization or vice versa the majorization-minimization (MM) algorithm open a wide avenue for optimizing the complex and high-dimension regularization models in a feasible way.
Theme 2: Environments that constitute soil, climate, landscape, and management practices are crucial for farmers to sustain their agricultural productivity. With the latitude and longitude information of the given farm locations, the NASA POWER Project and the Google Earth Engine provide daily/weekly/monthly historical weather data including temperature, precipitation, and so on. They are generally massive, high dimensions and exhibiting temporal-spatial correlations among data and therefore challenging to model. The proposed research aims to bridge the gap by addressing the lack of a mature transformer and information embedding model for environmental data and genomic data related to crop yield and cultivar breeding selection.
My proposed research in Theme 1, based on the MM algorithm, directs a new path toward optimizing the complex regularized regression models for various distributions, such as negative multinomial, mixture models, and zero-inflated Poisson, among others. This research is expected to have a significant impact on the development of methodology in statistics. In Theme 2, we take deep learning approaches in modelling environmental and genomic data for predicting cultivar performance. In both themes, the research outcomes will be used by Canadian farmers to modify their management practices, with the goal of promoting bee gut health and winter survival, as well as strategically selecting the most adapted cultivars for their farms. Overall, our research enables to sustain and increase agricultural production, and increase the diversity and economic equality in Canada's food production system.
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-2024-04982
Name nom
Feng, Zeny
Department département
Mathematics and Statistics
Organization id
24
Institution établissement
University of Guelph
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
34000
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
401
Area of application group en
Environment
Area of application group fr
Environnement
Area of application en
Climate and atmosphere
Area of application fr
Climat et atmosphère
Research subject code
3000
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Statistics and probability
Research subject fr
Statistique et probabilité
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 and computational methods for environment, life, and health.
Keywords
metagenomic microbiome; compositional data; statistical bioinformatics; environmental data; regularization; machine learning; neural network; embedding method; mixture models; mm algorithm
Application summary
Statistical and computational tools are indispensable for supporting research and tackling real world challenges in the fields of environment, life, and health sciences. My long-term goal is to develop and advance statistical and computational tools to address issues and analyze data arising from cutting-edge research in these disciplines. My research program has two main themes: 1) statistical methodology development for microbiome data analysis, and 2) environmental and climate data modelling.
Theme 1: The rapid growth of microbial metagenomics is attributable to the emergence of next generation sequencing. Massive amount of metagenomic sequencing data are available for exploring and disentangling the relationships between biological/environmental factors and gut microbiome compositions, and their impacts on host health. The proposed research program provides novel statistical methodologies in regularized regression models that overcome some of the known challenges existing in the field and provide effective and practical statistical tools for analyzing microbiome composition data. The proposed methods based on the formulation and utility of minorization-maximization or vice versa the majorization-minimization (MM) algorithm open a wide avenue for optimizing the complex and high-dimension regularization models in a feasible way.
Theme 2: Environments that constitute soil, climate, landscape, and management practices are crucial for farmers to sustain their agricultural productivity. With the latitude and longitude information of the given farm locations, the NASA POWER Project and the Google Earth Engine provide daily/weekly/monthly historical weather data including temperature, precipitation, and so on. They are generally massive, high dimensions and exhibiting temporal-spatial correlations among data and therefore challenging to model. The proposed research aims to bridge the gap by addressing the lack of a mature transformer and information embedding model for environmental data and genomic data related to crop yield and cultivar breeding selection.
My proposed research in Theme 1, based on the MM algorithm, directs a new path toward optimizing the complex regularized regression models for various distributions, such as negative multinomial, mixture models, and zero-inflated Poisson, among others. This research is expected to have a significant impact on the development of methodology in statistics. In Theme 2, we take deep learning approaches in modelling environmental and genomic data for predicting cultivar performance. In both themes, the research outcomes will be used by Canadian farmers to modify their management practices, with the goal of promoting bee gut health and winter survival, as well as strategically selecting the most adapted cultivars for their farms. Overall, our research enables to sustain and increase agricultural production, and increase the diversity and economic equality in Canada's food production system.