Generation and network analysis of an RNA-seq transcriptional atlas for the rat

Abstract

The laboratory rat is an important model for biomedical research. To generate a comprehensive rat transcriptomic atlas, we curated and down-loaded 7700 rat RNA-seq datasets from public repositories, down-sampled them to a common depth and quantified expression. Data from 590 rat tissues and cells, averaged from each Bioproject, can be visualised and queried at http://biogps.org/ratatlas. Gene correlation network (GCN) analysis revealed clusters of transcripts that were tissue or cell-type restricted and contained transcription factors implicated in lineage determination. Other clusters were enriched for transcripts associated with biological processes. Many of these clusters overlap with previous data from analysis of other species whilst some (e.g. expressed specifically in immune cells, retina/pineal gland, pituitary and germ cells) are unique to these data. GCN on large subsets of the data related specifically to liver, nervous system, kidney, musculoskeletal system and cardiovascular system enabled deconvolution of cell-type specific signatures. The approach is extensible and the dataset can be used as a point of reference from which to analyse the transcriptomes of cell types and tissues that have not yet been sampled. Sets of strictly co-expressed transcripts provide a resource for critical interpretation of single cell RNA-seq data.

Intraperitoneal transfer of wild-type bone marrow cells in theCsf1rknockout rat repopulates resident tissue macrophages without contributing to monocytopoiesis

“…[2] and there are currently no definitive markers. However, FLT3 ligand (Flt3lg) is highly expressed by rat T cells ( [37]; biogps.org/ratatlas) so it is likely that these CSF1R-independent populations are maintained by FLT3 signalling.…”