Genetic Architecture of AD: Update

[J11]

Great article! Will need to read the fine print, though, as they introduce a fair number of new concepts
to understand.

The article gives a good idea of where we are now in unraveling the genetic
architecture of AD. The snp heritability of AD is fairly low, yet it is still disappointing
to see how low the sample sizes and how close to little of this heritability has been found to date
(The blue dot to the above and to the left of the number 5 on the x-axis -- That's AD!! Right around 0?? Not sure
if that totally makes sense. Don't the current ~40 AD snps tell us anything?). The figure shows that
almost 100% of the genetics of snp AD heritability would be known at a sample size of ~5 million.
Yet, AD snp heritability is only ~8%. The contribution from apoe epsilon 4 is much larger than that
from other snps. It would be helpful if an updated population attributable fraction for known
AD snps were to be published.

Considering the improvement that would be possible in clarifying the genetic nature
of the illnesses shown in these figures it is not clear to me why larger studies have not been done.
Illnesses such as schizophrenia are moderately heritable and yet the research has not ramped
up as much as one might expect. With the imminent arrival of population scale full genome sequencing
perhaps We the People can push this over the line.

pgen.1008612.g003.jpg

pgen.1008612.g001.jpg

https://pubmed.ncbi.nlm.nih.gov/32427991/

[anotherdreamer]

I attempted to read the full article here: https://journals.plos.org/plosgenetics/ ... en.1008612 but a lot of the science went over my head. Since you clearly have a better grasp of this, would you mind answering a question?

This is from the article:
"For Alzheimer’s disease, our estimate of the liability-scale SNP heritability for the full 2013 dataset [39] is 15% for prevalence of 14% for those aged 71 older, half from APOE, while the recent “M2” and “M3” models of Zhang et al [14] gave values of 7% and 10% respectively–see S1 Appendix (p. S13)."

Does this mean that they believe only about 15% of AD is from genes? And that only half of that is from APOE? Or am I completely misunderstanding this?

[floramaria]

Great article! Will need to read the fine print, though, as they introduce a fair number of new concepts
to understand.

I attempted to read the full article here: https://journals.plos.org/plosgenetics/ ... en.1008612 but a lot of the science went over my head. Since you clearly have a better grasp of this, would you mind answering a question?

This is from the article:
"For Alzheimer’s disease, our estimate of the liability-scale SNP heritability for the full 2013 dataset [39] is 15% for prevalence of 14% for those aged 71 older, half from APOE, while the recent “M2” and “M3” models of Zhang et al [14] gave values of 7% and 10% respectively–see S1 Appendix (p. S13)."

Does this mean that they believe only about 15% of AD is from genes? And that only half of that is from APOE? Or am I completely misunderstanding this?

I can't weigh in o this myself because I don't understand it. I am re-posting anotherdreamer's question while also quoting J11 , so J11 will be notified.
anotherdreamer, when you are hoping for a direct response from someone who post you are responding to it is helpful to use the quote function. You can find info on this and other helpful hints for navigating the website most effectively at How-To Get the Most out of the ApoE4.info website

[anotherdreamer]

anotherdreamer, when you are hoping for a direct response from someone who post you are responding to it is helpful to use the quote function. You can find info on this and other helpful hints for navigating the website most effectively at How-To Get the Most out of the ApoE4.info website

Thank you, Floramaria! I didn't realize you had to be quoted to get an alert.

[floramaria]

Thank you, Floramaria! I didn't realize you had to be quoted to get an alert.

You are welcome!

[J11]

Yes, I also found the terminology somewhat confusing. Often what happens in science is a vocabulary or visual style develops and when researchers stray from the familiar people are not happy. The article introduces a fair number of newish ideas that makes it difficult to understand.

https://www.sciencedaily.com/releases/2 ... 232300.htm helps to clarify things. Alzheimer's is highly genetic. The url speaks of 80% of Alzheimer's being genetically caused. This might not even include epigenetic features. ~45% of identical twins are concordant for AD.

" ... folks who would have formerly been called sporadic, and testing how important genetic influences are ... and we're finding genetic influences are tremendously important"

{I wonder what the APOE 44 concordance is for identical twins? Most of the existing research speaks of AD or anxiety etc. as some sort of monolithic disease. Yet, clearly this is untrue. Researching this question could be of especial importance because it is entirely possible that some non-apoe4 AD could be highly non-concordant. What if an AD genotype could be found that had ~10% twin concordance? Would this not imply that a strong environmental exposure pushed one twin to dementia and not the other? I suppose the problem with such research would be that they would all be n=1 experiments because AD is so polygenic; APOE epsilon 4 is the only common strong AD risk allele.

I think a better way of thinking about this is that there is genetic risk for dementing illness.
Some people like the 45% concordant identical twins have a truly extreme genetic risk for AD.
For those discordant twins where one develops AD there might be some environmental influence diet, environmental toxin etc. that pushes them into dementia. Yet, for others the risk for dementia is probably almost non-existant. Over a wide range of potential environments they probably would not develop AD.

A dementia free future through genetic engineering is a near certainty.



I am becoming so excited about this new era of genetics!
We are transitioning to a post-human future.
Those at the 10% lowest risk of AD probably are essentially immune from dementing illness.
CRISPRing in the APP protective variant would create dementia free humans.

Embryo selection should allow us to select out all illnesses.
It is such a profound moment for humanity.

Full genome sequencing is now approaching $100.
This is a massive massive development.
It unlocks who we are and why we are who we are.
I am getting ready to go full genome; it is a step towards a transcendent future.
(see my new Full Genome thread.)

[floramaria]

Some people like the 45% concordant identical twins have a truly extreme genetic risk for AD.
For those discordant twins where one develops AD there might be some environmental influence diet, environmental toxin etc. that pushes them into dementia. Yet, for others the risk for dementia is probably almost non-existant. Over a wide range of potential environments they probably would not develop AD. ...
A dementia free future through genetic engineering is a near certainty. ...
I am becoming so excited about this new era of genetics!
We are transitioning to a post-human future.

I always enjoy reading your posts, J11, even though I do not share your enthusiasm for the post-human future!
Before we get to the stage you foresee where genetic engineering will create dementia -free world , epigenics factors are what we can optimize to minimize risk, yes?
You point out that for the discordant twins environmental influences would be important determinants. Since it not just individuals who carry the ApoE4 allele who develop dementia, it seems like a bit of a stretch to claim that genetic engineering alone will eliminate AD if epigenetic factors like toxicity and insulin resistance are not addressed.

[J11]

Thank you flora, I appreciate being appreciated. Yes, I actually also have qualms about the future that we are rapidly approaching. Often those at the front of the parade are every bit as concerned as everyone else, though they try their best to act with confidence. The people who have the greatest insight into the implications of genetic engineering appear to have been quite quiet about what is possible even now. We are moving towards entirely uncharted territory.

30 points of causal IQ has already been published. How could our species possibly survive such a profound uplift in intelligence almost overnight? Embryo selection already can give an average intelligence enhancement of 2.5 points. Even 2.5 points on a population scale would create extreme social change. Such change could happen every generation possibly for centuries.

One of my most cherished wishes has been to live in a world that was smarter. However, now that the research is exponentially ramping up, I am already becoming worried. Increasing human IQ by 30 points would intensify technological change by a factor of 100. A mere 90 points (which does not seem entirely implausible) would intensify "progress" by a factor of 1 million. How much progress can we live with before we are completely swamped?


Flora, what I have begun to appreciate is how genetics could potentially be used to override the environment. The APP variant is a good example. This rare variant largely would prevent AD under most normalizing scenarios. Simple genetic selection will allow for rapid social change. It can be so frustrating for those trying to help others move to better health when advice is often ignored. I am sure we would be all so much healthier if we were to lose 5-10 pounds, develop better sleep behavior etc.. If only we had behavioral modification technologies that were as effective as pharmaceuticals. Genetics gives us the option of simply checking off a box on a genetic engineering menu without having to make the difficult choices that most people ultimately are unable to live by. The future will be a place where people can make bad choices and not have to live with the consequences. (This is probably not entirely realistic, though it will be a powerful marketing selling point. You can have more ice cream and more chocolate and no down sides. This will be an unbeatable selling pitch.}

That is where the power of genetics enters the scene. All that will be needed is a one time change and life will be forever different. We could have a million year evolutionary leap in a single generation. No endless futile arguments begging people to stop smoking. Genetic engineering will create transformative change for our species.

I had thought of genetics and environment more as an either or. Yet, it can be more genetics and environment. One example that I found quite striking was that it was recently reported that a significant fraction of the people with schizophrenia in India have a niacin deficit. One could then say "well, niacin is an environmental cause of their illness". Yet, what is even more interesting is that
schizophrenic illness emerges in those with a niacin deficit most often with a specific variant in the NAD pathway. Genetically selecting against this variant would have a significant effect on schizophrenia risk in certain populations.

From this my expectation is that a truly profound reduction in disease risk could occur in the population over even the near term as genetic selection begins. Selecting embryos at the 10th percentile of risk (versus 90%+) for AD, schizophrenia etc. likely will have an extreme effect. The research that I have read on the genetics of schizophrenia noted that identical twins have ~50% concordance, fraternal twins have ~ 10% concordance, ordinary siblings have ~ 5% concordance, and cousins would be expected to have even less concordance. And of course this is in the context often of high risk multi-index schizophrenic families. The overall population risk is ~1%. Without even trying too hard it seems realistic to expect that future schizophrenic risk could be reduced perhaps >100 fold.
Such very large reductions should reasonably be expected across the board in human illness (including AD).

[circular]

How much progress can we live with before we are completely swamped?

I’m already there

I think gene editing will be like almost all other technologies, both good and evil in the combined hands of this existential soup we call the human species. Ohm.