The coding region of a gene is the part of the gene that will be eventually transcribed and translated into protein, i.e., the sum total of its exons. The remaining portion of the gene is interspersed by introns (see the figure below), or regions that are trimmed away during RNA splicing and thrown out. Meanwhile, the gene's exons are assembled into a single strand of mRNA for translation into protein.
Normally, only a small percentage of a given gene will actually form its coding regions. In fact, only about 5% of the human genome belongs to coding regions; in some eukaryotes, this percentage is as low as 1%.
Coding regions often correspond to GC-rich areas of the genome, or areas where the ratio of transitions to transversions is elevated (because transitions are less likely to change the encoded amino acid, which can cause problems in protein translation).
