P4‐103: Cell‐type Specific Mechanistic and Directional Transcriptional Regulatory Networks in Alzheimer's Disease
Year: 2019
Type: article
Abstract: Single cell RNA-seq is rapidly redefining our understanding of cell types in the brain. Limited in the number of genes detected per cell, single cell RNA-seq is too sparse for creating transcriptional regulatory networks (TRNs), which require quantitative expression for all transcription factors and target genes. Bulk tissue RNA-seq provides appropriate coverage of genes for TRNs, but is not cell-type specific. Using single cell RNA-seq from the Allen Brain Atlas, we have redistributed read counts in the AMP-AD bulk RNA-seq for cell-types of the brain, building cell-type specific TRNs. These TRNs represent a directional and mechanistic list of putative transcription factors for nearly all expressed genes in the brain. We reprocessed all ENCODE brain DHS using Wellington and HINT to identify transcription factor binding sites. We assembled motifs from multiple databases, totaling 1,530 motifs mapping to 1,515 transcription factors. We developed and utilized Transcriptional Regulatory Network Analysis (TReNA), as an R Bioconductor package. Regulatory regions were defined by GeneHancer. TReNA utilizes multiple machine learning techniques to prioritize transcription factors for each target gene. Single cell RNA-seq data from the Allen Brain Atlas was used to define cell types, for redistributing the reads from bulk RNA-seq. Our TRNs contain a prioritized list of putative transcription factor regulators for all genes in a cell-type specific manner. This information, combined with the footprints (and their genomic locations) allows us to integrate genetic information from GWAS or eQTL analysis to generate testable hypotheses around the functional annotation of variants. These cell-type specific TRNs can be used to identify key transcription factors enriched in gene lists identified in independent perturbation experiments of comparable cell types. We have identified multiple microglia-enriched transcription factors that regulate many differentially and co-expressed genes in AD. These resulting models can be applied to other datasets that generate lists of differentially or co-expressed genes, including single cell RNA-seq data, as well as provide testable hypotheses for non-coding variants of interest. We are actively engaged in testing several hypotheses through experimental means and have made these TRNs publicly available.
Source: Alzheimer s & Dementia
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: bronze
OpenAlex ID: W2980900416