Figure 2
From: An expression atlas of human primary cells: inference of gene function from coexpression networks
Clustering of samples based upon their gene global expression profiles. A Pearson correlation matrix was prepared comparing the data derived from all samples used in this study. A graph was then constructed using only those sample-to-sample relationships where r ≥ 0.9. Nodes represent samples and edges are coloured on a sliding scale according to the strength of the correlation (red, r = 1.0; blue, r = 0.9). The graph was then clustered using an MCL inflation value of 2.2, each cluster of samples being assigned a different colour. It is quite striking that almost without exception related cell types cluster together or are positioned within similar network neighbourhoods irrespective of the source of the data. Full details of the sources of all data sets used in the analysis is provided in Additional file1: Table S1. Additional abbreviations used: CMP, common myeloid progenitors; GMP, granulocyte monocyte progenitors; HSC, haematopoietic stem cell; mac., macrophage; MDDC, monocyte-derived DC; MDM, monocyte-derived macrophage; MEP, megakaryocyte–erythroid progenitor cell; MSC, mesenchymal stem cells; NK, natural killer; PB, peripheral blood; stim., stimulated; Treg, regulatory T cell.