The high-order chromatin structure plays a non-negligible role in gene regulation. However, the mechanism for the formation of varied chromatin structures in different cells and the sequence dependence of this process remain to be elucidated. For human and mouse genomes we identified CpG island (CGI) rich and poor genomic domains, which divide the genomes into two sequentially, epigenetically, and transcriptionally distinct regions. These two types of megabase-sized domains resemble the two homopolymer subunits in a block copolymer. They spatially segregate, but to a different extent and with different contents in different cell types. Overall, the two types of genomic domains gradually segregate from each other in development, differentiation, and senescence. The multi-scale spatial intermingling between the different domains is cell-type specific and increases in differentiation, thus helps to define the cell identity. We propose that the phase separation of the 1D mosaic sequence in space, serving as a potential driving force, together with cell type specific epigenetic marks and transcription factors, shapes the chromatin structure in different cell types and renders their distinct genomic properties. The mosaicity of genome differ for different species and could be related to their different biological behaviors such as body temperature control and senescence.