.png){kind=icon}
Return
Identifying gene expression programs in single-cell RNA-seq data using linear correlation explanation
.png){kind=icon}
PRE
AI
DOI：10.1016/j.jbi.2024.104644.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Objective: Gene expression analysis through single-cell RNA sequencing (scRNA-seq) has revolutionized our understanding of gene regulation in diverse cell types, tissues, and organisms. While existing methods primarily focus on identifying cell type-specific gene expression programs (GEPs), the characterization of GEPs associated with biological processes and stimuli responses remains limited. In this study, we aim to infer biologically meaningful GEPs that are associated with both cellular phenotypes and activity programs directly from scRNAseq data. Methods: We applied linear CorEx, a machine-learning-based approach, to infer GEPs by grouping genes based on total correlation optimization function in simulated and real-world scRNA-seq datasets. Additionally, we utilized a transfer learning approach to project CorEx-inferred GEPs to other scRNA-seq datasets. Results: By leveraging total correlation optimization, linear CorEx groups genes and demonstrates superior performance in identifying cell types and activity programs compared to similar methods using simulated data. Furthermore, we apply this same approach to real-world scRNA-seq data from the mouse dentate gyrus and embryonic colon development, uncovering biologically relevant GEPs related to cell types, developmental ages, and cell cycle programs. We also demonstrate the potential for transfer learning by evaluating similar datasets, showcasing the cross-species sensitivity of linear CorEx. Conclusion: Our findings validate linear CorEx as a valuable tool for comprehensively analyzing complex signals in scRNA-seq data, leading to deeper insights into gene expression dynamics, cellular heterogeneity, and regulatory mechanisms.
Keywords:
scRNA-seq
Machine learning
Single cell
Transfer learning
Developmental biology
Journal
IF:
4.5
Papers:
3.5K
Citations:
1.9W
Organization
View more organizations
Cited Papers
ShareSaveShareSaveMouse Genome Database (MGD): Knowledgebase for mouse-human comparative biology.png)
NUCLEIC ACIDS RESEARCH
OAAI
NUCLEIC ACIDS RESEARCHIF13.1
OAAIShareSaveThe dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cellsNATURE BIOTECHNOLOGYOAAI
NATURE BIOTECHNOLOGYIF41.7
OAAIShareSaveTracing colonic embryonic transcriptional profiles and their reactivation upon intestinal damageCELL REPORTSOAAI
CELL REPORTSIF6.9
OAAIShareSaveCellPhoneDB: inferring cell-cell communication from combined expression of multi-subunit ligand-receptor complexesNATURE PROTOCOLSOAAI
NATURE PROTOCOLSIF16
OAAIShareSaveDecomposing Cell Identity for Transfer Learning across Cellular Measurements, Platforms, Tissues, and SpeciesCELL SYSTEMSOAAI
CELL SYSTEMSIF7.7
OAAIShareSaveA single-cell atlas of liver metastases of colorectal cancer reveals reprogramming of the tumor microenvironment in response to preoperative chemotherapyCELL DISCOVERYOAAI
CELL DISCOVERYIF12.5
OAAIShareSave View more