ImmGen labs follow a rigorous process for data generation to maximize comparability between datasets of different origins. These protocols allow the derivation of reliable datasets from low numbers of highly purified cells directly ex vivo.

Most are generated from 5 week-old C57Bl/6J mice, shipped from the Jackson Laboratory one week prior to use (minimizing local environmental effects). In some specific cases, transgenic or KO mice are used as an alternative. Other inbred mice profiled also originate from the Jackson Laboratory (in collaboration with the Mouse Phenome Database project). Some use knockout mouse lines maintained in ImmGen labs (with matched littermates as controls).

All samples are sorted in ImmGen labs with a common Cell Preparation and Sorting protocol, with 2 successive sorts to maximize purity (most often 2 flow cytometry sorts), the final sort being directly into cell lysis on fixation buffer, depending on the protocol. These are then shipped to the ImmGen core team for processing using single microarray, RNAseq or ATACseq pipelines.

ImmGen Protocols:

- Cell Preparation and Sorting
- Total RNA extraction (Microarray)
- Poly-A+ RNA Sequencing (TruSeq)
- Ultra-Low-Input (ULI) RNA-seq
- Standard ImmGen 14-cell Set


ImmGen is updating the gene/transcript reference framework used to map and assign RNAseq data. Liang Yang has performed the herculean task of re-processing all existing ULI RNAseq data against GENCODE Version M25, from the previously used M16. This update applies to data on the “RNAseq Skyline”, "Population Comparison (RNASeq)" and "MyGeneSet" databrowser pages.
In addition, Gene tables for datasets deposited at NCBI’s Gene Expression Omnibus (GSE109125, GSE122108, GSE122597) have been updated to M25, as have the raw count numbers and normalized tables found on the ImmGen Dataset page.

Please be aware of small discrepancies that may occur while comparing to previously downloaded data or figures. Changes are minimal for most genes, but M25 differs from M16 by introducing around 2400 new genes and 8000 new transcripts (for a total of 55401 total genes, of which 21859 are recognized as protein-coding). Many of them are microRNA precursors, others correspond to better distinguished members of gene families. The most significant “newcomers” are Vps28, Tmem179b, and Itprip (primarily expressed in Mast cells, Macrophages and Granulocytes, respectively).