Distribution of spanion clusters around the transcription start sites of genes. The plot demonstrates the location preference of spanionclusters inside the fragments of the two Human databases. The transcript initiation site positions of the transcript proximal fragments are aligned to the zero position of the scale according to the ENSEMBL annotations. The position axis scale has no importance in the case of global sampling database. The spanioncluster content of the two databases (the area under the curves) is 0.29% and 5.2% relative to the full size of the databases respectively. The global sampling dataset represents the average spanionconcentration of the genome including intergenic regions and repetitive sequences as well. Mouse data are not shown.