Microglia, the major immune cells in the brain, mediate neuroinflammation, increased oxidative stress, and impaired neurotransmission in Alzheimer’s disease (AD), in which most AD risk genes are highly expressed. In microglia, due to the limitations of current single-omics data analysis, risk genes, the regulatory mechanisms, the mechanisms of action of immune responses and the exploration of drug targets for AD immunotherapy are still unclear. Therefore, we proposed a method to integrate multi-omics data based on the construction of gene regulatory networks (GRN), by combining weighted gene co-expression network analysis (WGCNA) with single-cell regulatory network inference and clustering (SCENIC). This enables snRNA-seq data and bulkRNA-seq data to obtain data on the deeper intermolecular regulatory relationships, related genes, and the molecular mechanisms of immune-cell action. In our approach, not only were central transcription factors (TF) STAT3, CEBPB, SPI1, and regulatory mechanisms identified more accurately than with single-omics but also immunotherapy targeting central TFs to drugs was found to be significantly different between patients. Thus, in addition to providing new insights into the potential regulatory mechanisms and pathogenic genes of AD microglia, this approach can assist clinicians in making the most rational treatment plans for patients with different risks; it also has significant implications for identifying AD immunotherapy targets and targeting microglia-associated immune drugs.

Traditional image genetics primarily uses linear models to investigate the relationship between brain image data and genetic data for Alzheimer’s disease (AD) and does not take into account the dynamic changes in brain phenotype and connectivity data across time between different brain areas. In this work, we proposed a novel method that combined Deep Subspace reconstruction with Hypergraph-Based Temporally-constrained Group Sparse Canonical Correlation Analysis (DS-HBTGSCCA) to discover the deep association between longitudinal phenotypes and genotypes. The proposed method made full use of dynamic high-order correlation between brain regions. In this method, the deep subspace reconstruction technique was applied to retrieve the nonlinear properties of the original data, and hypergraphs were used to mine the high-order correlation between two types of rebuilt data. The molecular biological analysis of the experimental findings demonstrated that our algorithm was capable of extracting more valuable time series correlation from the real data obtained by the AD neuroimaging program and finding AD biomarkers across multiple time points. Additionally, we used regression analysis to verify the close relationship between the extracted top brain areas and top genes and found the deep subspace reconstruction approach with a multi-layer neural network was helpful in enhancing clustering performance.