I have once again been hearing noises about the "thousand dollar genome" (TDG).  That is, a human genome read de novo for a USD 1000 or less.  Here (REVOLUTIONARY GENOME SEQUENCING TECHNOLOGIES -- THE $1000 GENOME), for example, is a request for proposals from the National Human Genome Research Institute to develop technology that would enable the TDG.

Based on my early efforts to quantify how the productivity and cost of sequencing were changing, Steward Brand asked me back in 2002 when we would get the TGD. 

Here is the plot I generated in response (click on the figure thumbnail for a full-sized version).

The cost hasn't changed dramatically recently, and at the current pace we we won't get the TDG until sometime after 2020.  With 3 billion (3x10⁹) bases in the human genome, we need to hit USD .3x10^-6 per base (which is .3 microbucks, 300 nanobucks, 300 nanodollars per base -- nanoeconomics anyone?) to reach the Thousand Dollar Genome.  However, the numbers on the plot are primarily based on instruments that use slab gel electrophoresis and capillary electrophoresis.  Thus as new technologies emerge we could very well get to the TDG much more rapidly.