SCRN: Single-Cell Gene Regulatory Network Identification in Alzheimer's Disease.

Area of Science:

• Neuroscience
• Genomics
• Computational Biology

Background:

• Alzheimer's disease (AD) is a leading neurodegenerative disorder with increasing prevalence.
• Gene regulatory disruptions in brain cells are implicated in AD pathogenesis.
• Understanding gene regulation is crucial for identifying AD mechanisms.

Purpose of the Study:

• To propose a novel graph learning method, Single-Cell based Regulatory Network (SCRN), for identifying gene regulatory networks in AD.
• To analyze single-cell RNA sequencing (scRNA-seq) data to uncover regulatory mechanisms in AD.
• To identify potential biomarkers and therapeutic targets for Alzheimer's disease.

Main Methods:

• Utilized UMAP dimension reduction for scRNA-seq data analysis of AD and normal samples.
• Applied the SCRN method, employing graph neural networks and heuristic link prediction.
• Constructed gene regulatory networks focusing on key AD genes (APOE, CX3CR1, P2RY12).

Main Results:

• Successfully constructed a gene regulatory network for Alzheimer's disease using SCRN.
• Enrichment analysis revealed significant pathways, including NGF signaling, ERBB2 signaling, and hemostasis.
• Demonstrated the feasibility of SCRN in uncovering AD-related biological processes.

Conclusions:

• The SCRN method is effective for identifying gene regulatory networks from single-cell data in AD.
• SCRN facilitates the discovery of novel biomarkers and potential therapeutic strategies for Alzheimer's disease.
• This approach offers insights into the complex gene regulation underlying AD pathogenesis.