Interesting commentary in Genome Research by Phil Green, of Phrap fame. He argues that whilst useful information can be gleaned from low-coverage genome sequence, we are surely missing the most interesting regions:
More seriously, since very few features of any appreciable size (e.g., genes) will be completely covered, analyses requiring complete features cannot be carried out. In addition, as was noted above, whole-genome assemblies (of any depth) often fail to incorporate a significant fraction of the repetitive sequence in the genome. This is often considered to be a relatively minor deficiency, which may be true so long as the primary research focus is on broadly shared biological features. However, it is now apparent that repetitive sequence is a key agent of evolutionary change: Segmental duplications are likely the primary source of new genes…and recent evidence strongly suggests that transposable elements are important mechanisms of regulatory innovation…As researchers’ attention turns toward understanding differences among organisms rather than similarities, inadequate coverage of these types of sequences will become increasingly problematic.
Article concludes: “A major function of the 2x genomes will no doubt prove to be whetting users’ appetites for more complete sequences.”