The value of complete microbial genome Sequencing (you get what you pay for)

In the article above, we present arguments in favor of complete genome sequencing as an important research tool. In the following article by Branscomb and Predki (J. Bacteriol. 184.6406-6409, 2002) about draft microbial genome sequence data, the authors raise the notion that "delayed, third-party, or targeted finishing can be made to work very efficiently as a second step to draft sequencing." We believe that it is too early to make such conclusions based on what has been accomplished to date. In fact, this contradicts another statement made by Branscomb and Predki that "it may be substantially more expensive on average to finish draft sequence data later,.. than to do so at the start and in the same laboratory." Given that cost savings is the primary reason for carrying out draft sequencing, we still strongly believe that except under unusual circumstances, such cost savings will be minimal. However, we agree with Branscomb and Predki on many other points, in particular the hidden costs that derive from (i) errors and imperfections in the data that may be misleading, (ii) the additional costs that come from uncoupling the finishing of draft sequence from its generation, and (iii) the risk that many of these projects will never be finished because of shifts in research priorities. We also agree that there is value in generating draft sequence data in some circumstances, but we maintain that complete genome data should be the target whenever possible.