Grant: Master's Award: Canada Graduate Research Scholarships - Master's program
University of Toronto
$27,000.00
Explainable Deep Learning Model for Gene Regulatory Network Inference from Single-Cell Multi-omics Data
EF 2025-26
Résumé
Understanding gene regulation is essential for explaining how cells differentiate and give rise to different phenotypes. These patterns of control are governed by gene regulatory networks, which determine how genes and their regulatory elements interact. New single-cell technologies now enable us to measure multiple layers of information at once, such as gene expression and DNA accessibility. These datasets offer an opportunity to understand how genes are regulated, especially when cells respond to stress and drug treatment. However, current computational tools often struggle to identify causal effects within these networks. Many approaches rely on correlations or fail to work well across different cell types. As a result, important regulatory interactions may be missed, and predictions often do not generalize to new biological contexts.
My research aims to address these challenges by developing an interpretable deep learning framework that can more accurately uncover how genes are controlled. I will begin by evaluating existing models on diverse single-cell datasets where regulatory elements have been experimentally perturbed. These analyses will highlight where current methods fail and will guide the design of improved models that incorporate biological knowledge and multimodal information. I will also use explainable AI techniques that show which DNA sequences or expression patterns drive each prediction. This project will advance our ability to map regulatory programs involved in immune responses, cell differentiation, and disease. By generating interpretable and biologically grounded predictions, it will support stronger experimental design, help identify important regulatory targets, and guide future laboratory studies.
Programme
Master's Award: Canada Graduate Research Scholarships - Master's program
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database
Tel que publié
Id
573039
Name
Xu Angela
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Explainable Deep Learning Model for Gene Regulatory Network Inference from Single-Cell Multi-omics Data
Programtype2
Studentship Programs
Orgnameinp2
University of Toronto
Cihramount2
$27,000
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Research Scholarships - Master's program
Keyworddelim
Computational Biology; Explainable Ai; Gene Regulatory Network; Genomics; Machine Learning
Abstract
Understanding gene regulation is essential for explaining how cells differentiate and give rise to different phenotypes. These patterns of control are governed by gene regulatory networks, which determine how genes and their regulatory elements interact. New single-cell technologies now enable us to measure multiple layers of information at once, such as gene expression and DNA accessibility. These datasets offer an opportunity to understand how genes are regulated, especially when cells respond to stress and drug treatment. However, current computational tools often struggle to identify causal effects within these networks. Many approaches rely on correlations or fail to work well across different cell types. As a result, important regulatory interactions may be missed, and predictions often do not generalize to new biological contexts.
My research aims to address these challenges by developing an interpretable deep learning framework that can more accurately uncover how genes are controlled. I will begin by evaluating existing models on diverse single-cell datasets where regulatory elements have been experimentally perturbed. These analyses will highlight where current methods fail and will guide the design of improved models that incorporate biological knowledge and multimodal information. I will also use explainable AI techniques that show which DNA sequences or expression patterns drive each prediction. This project will advance our ability to map regulatory programs involved in immune responses, cell differentiation, and disease. By generating interpretable and biologically grounded predictions, it will support stronger experimental design, help identify important regulatory targets, and guide future laboratory studies.
Competitiondate
202512
Approvedterm2
1 yr 0 mth
Pinamesdelim
Xu, Angela
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $27,000~ $0~