Slashdot Mirror
How To Check Yourself For Abnormal Genes
AnneWoahHickey writes "While the State of California was harassing personalized genomics companies, and hindering the development of personalized medicine, Wired was preparing a guide to genetic testing. It explains how to make sense of the massive sets of raw data offered by 23andMe or deCODEme, and a way to check yourself for genetic abnormalities that are not covered by microarray tests. Facing a medical community that is fiercely resistant to change, the fate of personalized medicine is truly in the hands of consumers."
OK - so first of all 23andme et al do not search for "abnormal genes" - they look for common polymorphisms present in human DNA sequences. These are not abnormal, simply different. Secondly, rs numbers found in association with disease are practically valueless without the underlying functional data, plus replication of the association in different populations. For Zeus' sake, bear this in mind if you ever get one of these tests!
No way in hell anyone who hasn't had massive experience with PCR is going to get results from a DIY PCR. Extracting DNA from a sample is dead easy with the latest generation of kits, and DNA Is fairly stable stuff, but PCR protocols, although simple, are incredibly touchy and take a lot of time to get consistent results from.
The rough equivalent of having a page that says to Joe Public that he can either pay some professional to build a custom database for his companies needs, or he can download OpenOffice and do it himself. It's only cheaper if you don't put a value on time, quality or results.
Yes, I know, I must be new here...
Thanks for the link. Since we're at it, I'll repost a link I posted in response to the thread a couple of weeks ago on the same subject.
The US Government Accountability Office compiled a report of genetic testing that is available here. I'm not posting any quotes from it but its quite strongly worded conclusions are that these online genetic tests are at best worthless and at worst harmful. Any government that doesn't try to shut them down is being negligent.
This article was clearly written by someone who has no clues to this kind of work. It covers the basic steps although the draft described would not even work (designing primers just by picking 20 bp sequences without checking if you design them into some repeat, or other non unique sequence, without checking that there is no hairpin formation, no primer dimers etc, also he just says 40 cycles in PCR machine without saying that for each prime pairs you need a specific annealing step and describing what other heating steps are required in the PCR machine). Other thing is that he forgot to mention costs and time to do this. Lets say a primer pair is just $1 (it is more even if you order the smallest amount) and one PCR run is roughly 2 hours (with 40 samples) also preparing 40 different samples takes like at least 1 hour of work. Plus you need the materials for PCR (PCR grade water, MgCl2, buffer, the polymerase ensyme, for like 100 reactions at cheapest you can buy them for like $50-100). The PCR machine cost will be almost negligible with its $1000. Now calculate the costs and time needed for like 1 million SNPS. And you realize that home made traditional PCR techniques won't work. Lastly what if you find some SNPs different than others. You need to know the different databases, you need to be able to filter the 99% junk from somethign valid since most of the SNPs are just variations without any change of the functionality. At best they are linked to some disease at a given population and could have no meaning at an other population.
Speaking as someone who has done a PhD on genome-wide microarray SNP analysis, I can tell you that we are not yet at a point of maturity where you can simply put a drop of blood in a machine and get reliable prognostic information or lifestyle and treatment recommendations.
The technology is actively researched, i.e. most often we're not looking at the results from a clinical standpoint but as an indicator of the performance of a certain method. Practically speaking, only research centers are interested at the stuff and you would be extremely hard pressed to convince practicing doctors to incorporate current results in their everyday work, even though some studies have appeared in famous medical journals (New England Journal of Medicine, Nature etc). Using software notation, the results are "alpha" grade at the moment.
That being said, there is no harm in knowing that you have an Adenine in position XXXX. Harm comes from acting upon that knowledge without sufficient clinical evidence.
P.
Whoever wrote this article shows a gross misunderstanding about how genetics actually works. The central dogma of genetics applies here: DNA is transcribed into mRNA, and translated into proteins, which can then be post-translationally modified.,
First - a single nucleotide change may or may not cause a "genetic defect." Translation involves taking three nucleotides (aka bases) and getting the appropriate amino acid from that. There are 20 common amino acids, and 64 combinations - so there is some overlap. If the changes nucleotide doesn't change the corresponding amino acid, it doesn't matter.
Second - not all mutations are harmful. If a mutation happens in an exon (a piece that is cut out), there may well be no difference if there is a mutation there or not. Even if it' is in a part that is kept, it may not be in a part of the protein that dictates structure or function.
Third - most organisms, including humans, have built in redundancies and backups. Losing a gene doesn't usually mean losing the protein, because often something else will make the product another way, or compensate. In diploid organisms often this can be duplicated genes or the other allele.
In short, in order to truly make sense of the data given by these companies you really need to know at least the basics of genetics and have an understanding of how the gene and protein work. These are no small tasks and, surprise, generally results in getting a degree in some branch of biology.
But when people still listened to The Offspring, CDs were popular.