Grant: Discovery Grants Program - Individual
University of Waterloo
$31,000.00
Uncertainty-aware integrative pattern discovery for single-cell analysis
FY 2024-25
Summary
Background:
Single-cell sequencing technology has enabled the high throughput interrogation of many aspects of genome biology. However, single-cell sequencing data has unique characteristics, related to noise and sparsity, and information about the resulting uncertainty in the data often fails to propagate to downstream analyses. Additionally, most single-cell assays aim to capture a single modality of the cell, limiting our ability to draw a more holistic picture of the relationships between biomolecules in single cells. Therefore, new scalable approaches are needed to enable joint analysis across multiple single-cell modalities, taking into account the presence of measurement uncertainty and batch effects.
Objectives:
The goal of the proposed program is the development of uncertainty-aware and interpretable methods to identify integrative patterns from noisy and heterogeneous single-cell data. As a first step toward this goal, I have developed a machine learning framework, called "Adversarial Clustering Explanation" (ACE), that aims to jointly model the whole single-cell RNA-seq pipeline, from embedding, clustering, to cluster explanations. ACE joint modeling enables the uncertainty in the nonlinear embedding and stochastic cluster assignments to be propagated to the downstream analysis. The proposed program aims to expand the capabilities and scope of ACE in following important ways: (1) Identify cell type-specific gene panels subject to false discovery rate control; (2) Identify gene panels that explain developmental trajectories in a continuous fashion; and (3) Identify integrated explanatory patterns from heterogeneous data modalities such as gene expression and chromatin accessibility.
Impact and long-term goal:
The proposed program will produce an open source, state-of-the-art single-cell analysis pipeline capable of jointly analyzing heterogeneous modalities of single-cell data, even in the presence of measurement uncertainty and batch effects. The computationally efficient, statistically rigorous, and scientifically meaningful methods tailored for highly complex biomedical data will potentially accelerate the research and development of biomedical research in Canada. More importantly, the identified novel cell type-specific patterns from the genome sequence, gene expression, and chromatin accessibility can inspire the identification of novel disease biomarkers, cellular subsets, therapeutic targets and diagnostics or guiding intervention strategies in potential therapeutic applications in Canada. Finally, as an important part of the program, students will learn the necessary skills to carry out a research plan, scientific writing, critical thinking, and coding. By the end of the proposed program, the students will become machine learning experts who are well-positioned for industrial research scientist positions or academic research positions in top universities.
Program
Discovery Grants Program - Individual
Recipient
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
RGPIN-2023-03270
Name nom
Lu, Yang
Department département
Computer Science
Organization id
33
Institution établissement
University of Waterloo
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
31000
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
1507
Committee name en
Computer Science
Committee name fr
Sciences informatiques
Area of application code
1201
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Life sciences (including biotechnology)
Area of application fr
Sciences de la vie (y compris biotechnologie)
Research subject code
2700
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Information technology
Research subject fr
Technologie de l'information
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
Uncertainty-aware integrative pattern discovery for single-cell analysis
Keywords
single-cell genomics; single-cell rna-seq; hypothesis generation; hypothesis testing; heterogeneous data integration; interpretable machine learning; pattern discovery; uncertainty; false discovery rate control; computational biology
Application summary
Background:
Single-cell sequencing technology has enabled the high throughput interrogation of many aspects of genome biology. However, single-cell sequencing data has unique characteristics, related to noise and sparsity, and information about the resulting uncertainty in the data often fails to propagate to downstream analyses. Additionally, most single-cell assays aim to capture a single modality of the cell, limiting our ability to draw a more holistic picture of the relationships between biomolecules in single cells. Therefore, new scalable approaches are needed to enable joint analysis across multiple single-cell modalities, taking into account the presence of measurement uncertainty and batch effects.
Objectives:
The goal of the proposed program is the development of uncertainty-aware and interpretable methods to identify integrative patterns from noisy and heterogeneous single-cell data. As a first step toward this goal, I have developed a machine learning framework, called "Adversarial Clustering Explanation" (ACE), that aims to jointly model the whole single-cell RNA-seq pipeline, from embedding, clustering, to cluster explanations. ACE joint modeling enables the uncertainty in the nonlinear embedding and stochastic cluster assignments to be propagated to the downstream analysis. The proposed program aims to expand the capabilities and scope of ACE in following important ways: (1) Identify cell type-specific gene panels subject to false discovery rate control; (2) Identify gene panels that explain developmental trajectories in a continuous fashion; and (3) Identify integrated explanatory patterns from heterogeneous data modalities such as gene expression and chromatin accessibility.
Impact and long-term goal:
The proposed program will produce an open source, state-of-the-art single-cell analysis pipeline capable of jointly analyzing heterogeneous modalities of single-cell data, even in the presence of measurement uncertainty and batch effects. The computationally efficient, statistically rigorous, and scientifically meaningful methods tailored for highly complex biomedical data will potentially accelerate the research and development of biomedical research in Canada. More importantly, the identified novel cell type-specific patterns from the genome sequence, gene expression, and chromatin accessibility can inspire the identification of novel disease biomarkers, cellular subsets, therapeutic targets and diagnostics or guiding intervention strategies in potential therapeutic applications in Canada. Finally, as an important part of the program, students will learn the necessary skills to carry out a research plan, scientific writing, critical thinking, and coding. By the end of the proposed program, the students will become machine learning experts who are well-positioned for industrial research scientist positions or academic research positions in top universities.