[thedoc (OP)]

Paul asked the Naked Scientists:
   
What is the best sample of DNA that I can make?
Spit in a bottle, a scraping from inside my cheeks, a patch of dead skin off my elbow, a snip of hair or what? If it is gray or white hair how will that impact on the quality of the DNA?

What do you think?

[evan_au]

Most DNA samples for legal reasons are taken from a buccal swab from the inside of the mouth, as it causes minimal inconvenience.

A better source of DNA is from white blood cells, but this requires taking a blood sample, so it is a bit more invasive, and it is more difficult to store and transport.

This site compares these methods against its commercial product (other commercial products are available!):
http://blog.dnagenotek.com/blogdnagenotekcom/bid/35944/Rinse-Swab-or-Spit-What-s-the-Real-Source-of-DNA-in-Saliva

For research into cancer genetics, biopsy samples are taken from multiple locations within the cancer. This is not simple or painless, and requires extensive analysis.

For the average person, there are a number of companies who offer a commercial genetic testing service to the public, and will provide a sample kit (eg here). Today, these commercial offerings don't scan the entire genome, but only analyse a small fraction of it which commonly varies between individuals, called "Single Nucleotide Polymorphisms".

However, different organisations were using the results of different research papers, and testing a different set of SNPs, and using a different computer program to analyse the probabilities, which resulted in a different calculation of the risks for an individual. Eventually the US FDA banned them from providing an analysis for the public, but they do provide resources so that an interested person can do some analysis for themselves. (Maybe the analysis is still available to people outside the USA?)

DNA sequencing technology has been progressing rapidly - with much faster improvements than "Moore's Law" which has revolutionized electronics for the last 50 years. For some time, researchers have talked about the "Holy Grail" of a DNA sequence costing less than $US1000; at this price, it would almost be affordable (at least in "rich" countries) to do a full DNA transcript for everyone at birth.

The main barrier is that while the cost of reading the DNA is likely to reach $1000 in the next few years, applying the skills to understand how this DNA may result in disease for the individual currently requires input from a large number of experienced professionals, at an estimated cost of around $50000. And if a DNA mutation has only been seen in one individual (ie it is not common in the community), very often the answer is "we don't know"!

And after that, even if a particular dangerous mutation is found, we often have no approved medication that can treat it. Developing treatments for "1-off" conditions will require a total overhaul of our current drug development and approval process, which is aimed at development of mass-market drugs which can be supplied to millions of people.

So DNA sequencing will not become standard practice until the cost of analysis plummets, and it can be shown that the cost savings from having this information will save far more than the cost of transcription and analysis. (Not to mention the potential social upheaval of paternity and ethnic analysis, which is a natural byproduct of global genetic screening.)