Razib Khan points to a projection for the rate at which full human genome DNA sequencing will be done over the next 10 years.

The firm GenomeQuest has a blog, and on that blog they have a post, Implications of exponential growth of global whole genome sequencing capacity. In that post there are some bullet points with numbers. Here they are:

* 2001-2009: A Human Genome

* 2010: 1,000 Genomes – Learning the Ropes

* 2011: 50,000 Genomes – Clinical Flirtation

* 2012: 250,000 Genomes – Clinical Early Adoption

* 2013: 1 Million Genomes – Consumer Awareness

* 2014: 5 Million Genomes – Consumer Reality

* 2015-2020: 25 Million Genomes And Beyond – A Brave New World

In light of the many orders of magnitude drop in DNA sequencing costs we've already witnessed in the last 30 years I find these numbers plausible. The 2010s are going to be a period in which the functional significance of tens or hundreds of thousands of genetic variants become known and DNA sequencing becomes affordable for the middle class.

There many reasons to do genome sequencing. Not all these reasons involve clinical applications by health care providers, though most do.

• Scientific discovery. The discovery drives the usefulness of sequencing for other purposes.
• Future disease prediction coupled with efforts to avoid your genetic fate.
• Genetic compatibility of drug choices. Some genetic variants make certain drugs harmful. e.g. a genetic variant that causes an antibiotic to cause liver toxicity.
• Mating choices. Choose mates or IVF embryos based on genetic tests.
• Career choices. Find out if you have the genes needed to excel in a sport or in an intellectual career or even other career choices.
• Self understanding. Find out why you are what you are and how much of who you are is due to your genes.
• Cancer treatments. Aim therapies at the mutations which a cancer has.

I expect DNA sequencing of cancer cells will become routine. Knowledge about cancer genome mutations will be used to choose treatments. I would not be surprised if DNA sequencing data gets used to design antibodies to attack cancer cells without attacking normal cells.

Since a person can have cancer for months or years and since cancer cells constantly mutate I expect oncologists to track cancers by using repeated sequencing of many cancer cell samples from each patient. Once full genome DNA sequencing costs just a few hundred dollars the cost will become a small fraction of total cancer treatment costs. So I do not see money as an obstacle to this style of usage of sequencing machines.