Lg1, yes you are quite right a clear summary of the facts as they are now known would be helpful.
The study found that the minor allele (that is, G) of rs9472817 in the SLC25A27 gene appeared to reduce AD risk in epsilon 4 carriers.
This study likely had few if any epsilon 44s, so it is very unclear whether the result applies to them.
For those without this SNP on their genechip, the workaround is the SNP rs9381468.
The minor allele (that is, G) of rs9381468 should equate to the G allele of rs9472817.
This should give good, though not perfect predictive power.
The minor alleles of rs9296504 (that is, C) and rs9381469 (that is, A) might help to improve the prediction.
We could go to OpenSNP or 1000 genomes to confirm that the accuracy of this approach.
At this point the result needs to be treated as highly speculative, there have been thousands of previous such claims that have
nearly without exception been subsequently proven to be completely wrong. Even if the thin chance that this is actually true holds,
it would be expected that the true benefit would be much less than has been reported.
The 80% of people who have been made miserable by this news really should cheer up!
As I have mentioned, it probably isn't correct.
Also, my objective assessment is that a defect such as this seems fixable.
Many genetic variants (for example, epsilon 4) have been studied for decades and it is still not clear what the underlying biological
problem is. The biology of Uncoupling Proteins seems much simpler. These proteins uncouple the mitochondrial proton motive force: Simple!
(Anyone care to give a one sentence biological description of what APOE does?)
I see this more as a great learning moment.
First, we clearly need a freely accessible database of AD genotypes and phenotypes.
The researchers of the current study had to go on a hunch and gamble a fair chunk of change to try and prove their point.
This puts a substantial barrier to solving the genetics of Alzheimer's. Whoever authorized the spending must have realized that this
had lottery odds of paying off. If they could have run this first through a database, then they would have significant confidence that it would work out for them. It is quite surprising from a financial perspective that this is not done. It is taxpayer's money: our money.
Why isn't there an AD database? Don't the Alzheimer's patients who donated their DNA samples for this research at some level "own" their
DNA and the informational value that is locked up in restricted access computer servers? Should not their rights concerning possible disclosure of their data be respected. We would not have to go through this somewhat traumatic process of sorting through all these variants if we could now access such a database. The truth about the 817 SNP could be known immediately.
We need an open AD database.
Secondly, this also highlights for me the importance of people knowing their genetics. There are quite a few out there that must be hoping that they were GG. Once AD genetics has been definitively unraveled, there were quite likely be a variant ( rs9472817 or otherwise) in which people will wish that they had it. I know how tough it would be in this age of genetics knowledge to look at one's child and realize that by some unfortunate accident they got the wrong allele. If enough people were aware of being in this circumstance (by being genotyped), then we really could take the future into our own hands.
The answer is of course obvious.
Why not just change the genetic mistake?
Why not CRISPR?