This directory contains data files and scripts related to our work: "Gene expression variability across cells and species shapes innate immunity".
In this work, we have studied the relationship between transcriptional divergence between species in response to immune challenges and expression variability between individual cells. We used two cross-species systems to study this: dermal fibroblasts and bone marrow-derived mononuclear phagocytes.
We further study the relationship between these variability and divergence and other characteristics - including:
- promoter architecture (presence or absence of elements such as CpG islands or TATA-boxes),
- gene functions, and
- other evolutionary and cellular characteristics (such as coding sequence evolution, number of cellular interactions).
The characteristics used are always related to one reference species: human - in the case of fibroblasts, and mouse - in the case of phagocytes. For example, when studying the relationship between cross-species divergence and promoter architecture, we use human and mouse promoters, for fibroblasts and phagocytes, respectively.
The scripts in this directory allow reproducing the statistical analysis and main plots. We also provide a script detailing how we compute cell-to-cell variability using the DM method, and pipelines for the fibroblast ChIP-seq and phagocyte single-cell analyses. Larger files and raw data (such as fastq files and count matrices) can be found in our FTP website and the relevant accessions in ArrayExpress (see manuscript for details).
Most groups were compared using Mann-Whitney test. Similar results are obtained using alternative tests. The code can be changed to output p-values computed using different methods by changing a flag in common_functions.R