Genetic variant accelerates normal brain aging in older people by up to 12 years | Hacker News Reader

Genetic variant accelerates normal brain aging in older people by up to 12 years

neurosciencenews.com

101 pointsbaalcat57 comments

Note: The following is complete speculation and probably bullshit. Please correct me!

I can't find the actual paper for this one. But reading an older study [0], also about TMEM106B, it seems they had already established an association between three SNPs and frontotemporal lobar degeneration (FTLD) risk.

However, the surprise discovery back then seemed to be the large discrepancy between the controls and the subgroup FTLD-GNR (those with FTLD and GNR mutations) for TT rs6966915 and CC rs1990622. See table 2 in [0], and look at the odds ratios. They are remarkably low for TT/CC, which invites further study that may lead to understand how to protect against FTLD (by understanding possible protective mechanisms, even therapies and so on).

As for listing out those odds ratios for your 23andMe genome, you can do it with arv [1]. For table 1 in the study (unless you know you have GNR mutations):

  import arv
  
  genome = arv.load("genome.txt")
  
  rsid = "rs6966915"
  gt = genome.get_snp(rsid).genotype # plus orientation
  print("%s %s" % (rsid, arv.unphased_match(gt, {
      "CC": "CC - OR 0.94",
      "CT": "CT - OR 1.04",
      "TT": "TT - OR 0.74"})))
  
  rsid = "rs1990622"
  gt = ~genome.get_snp(rsid).genotype # minus orientation
  print("%s %s" % (rsid, arv.unphased_match(gt, {
      "TT": "TT - OR 0.93",
      "CT": "CT - OR 1.04",
      "CC": "CC - OR 0.74"})))

Again, did I say that I'm a complete noob? Be very careful drawing conclusions from the program (or believing I know what I'm talking about --- I don't!)

[0]: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3034409/

[1]: https://github.com/cslarsen/arv

Welp, looks legit to me.

Someone who knows what they're talking about tell us noobs any major gotchas with this sort of approach, please :)

I wonder whether this variant is similar to APOE E4 allele—which significantly increases one's risk of Alzheimer's—in that its affect can be minimized by lifestyle changes.

In men, APOE E4's disadvantages can be almost completely eliminated (compared to APOE E3) by regularly exercising and completely abstaining from alcohol.

As I understand this only works for E3/E4 variants. Not homozygous E4/E4.

I haven't seen many research papers that claim abstaining from alcohol helps reduce ALZ risk in E4 carriers. Care to share?

Looks like the main SNP in consideration is: https://www.snpedia.com/index.php/Rs1990622. Anyone see the particular allele value that increases risk? My guess (looking at some genetic family data) is if you having an "A" instead of a "G" is the mutation - can anyone confirm?

Looks like that's it: "We identified 2 coding variants: one rare variant which was observed in a single control (c.401G>A; p.S134N) and one common variant (rs3173615, pT185S)." (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3034409/)

Heads-up: 23andme data contains this gene:

cmbpr:raw_23andme caleb$ grep rs1990622 *.txt

genome_Person1.txt:rs1990622 7 12283787 GG

genome_Person2.txt:rs1990622 7 12283787 GG

genome_Person3.txt:rs1990622 7 12283787 AG

I have 'A or G'

https://itsosticky.com/1qz58aj

Any idea what this means?

I hope someone who understands the article can offer an explanation about whether it's possible to use the 23andme data to speculate about one's own susceptibility to this.

Seconded.

The mutations found are in the protein TDP-43 [1]. This protein is also involved in ALS. It appears to be a regulatory DNA-binding protein. The protein also helps splice the mRNA of CFTR - the cystic fibrosis transmembrane conductance regulator.

[1] http://www.uniprot.org/uniprot/Q13148

As I've gotten older I've noticed my memory is not as good (both short term - what was I talking about, and long term - what was the very obvious term for this thing I've said many times in the past).

Any other middle-aged people (early 40s) experience this? What degree of it is just "your brain sucks as you age"?

I'm 29 and I've been experiencing this all my life. Especially the very obvious word part. That's how "floor spatula", "floor blanket", "trash panda", "majestic sea flap flap" and similar fun terms are born. (some of those are not mine)

Short term memory lapses I mostly attribute to inattention and mind racingness. But I'm not old yet so maybe it's different.

Is "floor spatula" a dustpan?

Yes. See, it works :)

I love it! I have a tendency to do the same thing and I suspect it'll only get worse as time goes on (or better, depending on how you look at it). I'm definitely stealing floor spatula though.

It's quite frustrating to be able to come up with very complex descriptions and not remember the obvious name/word.

Actual thing I've said: "What's the name of that actor that played Han Solo, Indiana Jones and who recently nearly crashed his own plane into another full of passengers by landing on a taxiway?"

I don't really believe young brains are that good either. I'm young and it's not like I remember a lot of things. When I practice something I see the improvement in that area. When I don't I forget it.

What I mean is I could complain all day.

I have found that times in my life when I exercised less, I forgot more.

I also find less time to exercise as I grow older.

However, for the same level of exercise, rest, and study, I can't tell if there is a difference.

I also have more to forget, so I can barely remember entire arguments in philosophy that I had down cold 10 years ago, but even at age 20, you may not remember much from when you were 10.

That's a good point, about how when you're 20 you hardly remember much from 10+ years ago.

In that sense, my memory is excellent, now I can remember a lot from 20 years ago.

Yep, definitely aren't as sharp as when I exercise more. And come to think of it, there was a good 13 year period where I was cycling a lot just to get around and now I drive to work every day.

I don't think having a smartphone in my pocket helps either, I tend to pull it out and read HN for 30 second periods during the day, then lose track, somewhat, of where-and-what I was doing.

"I have found that times in my life when I exercised less, I forgot more. I also find less time to exercise as I grow older."

Want to improve your memory and brain function at any age? Exercise. [0] Want to add new neurons (neurogenesis), exercise. [1] Want to improve your neuro-plasticity? [2] Learn new skills.

It's always bothered me nerds/tech-heads optimise their lifestyles for time, not realising fitness, novelty and rest can be used as a strategic weapon to getting better at what you do. I'm not being critical of the OP, I'm just observing what I see in my own tech circles.

Reference

[0] http://www.health.harvard.edu/blog/regular-exercise-changes-...

[1] https://www.dnalc.org/view/848-Exercise-induced-Neurogenesis...

[2] http://www.nature.com/nature/journal/v427/n6972/full/427311a...

I believe my memory, both short term and long term, isn't as good as it once was. I'm 36.

I sometimes wonder how much of this is due to stress, tiredness, inattention, disinterest. All those things probably contribute. But then I go on to think: no, this is just what happens when you age. One day you will not be able to hack it anymore and die.

I think part of forgetting as you're older is simply that there's less novelty in your experiences. You're going to remember your first 100 movies way better than your 1000th.

Your first ten trips to Red Robin (tm) vs your 40th. And so on.

It can also be related to stress or lack of sleep. Both of these things are absolute memory killers - to a point where people suffering from it literally worried about dementia.

> “TMEM106B begins to exert its effect once people reach age 65,”

Genes does not work that way. Unless a specific protein and particular pathway are identified, and the effect of mutated protein is validated, this is mere a speculation.

That's only colloquial, not wrong. The corrolation between this gene and the observed effects exists only after age 65. I don't see that this precision would change much in the way the intended reader is going to interpret it. Something about the presence of this gene[1] is causing early frontal cortex degeneration after age 65.

[1] Unlike lots of other situations in medicine, the direction of causality is known here: this gene obviously can't be the result of some other age-related driver of degeneration. Genes do not work that way, heh.

Cant wait till insurance drops people based on this gene

I know it's a joke, but it's important to point out for those who don't know that the ACA as it stands right now disallows discrimination on the basis of any pre-existing condition (which obviously includes genetic makeup), and even the AHCA disaster currently failing to pass in the house doesn't change that.

The dystopia might arrive someday, but not now, and not in the near future. Democrats (yes, it was all democrats) got it done seven years ago.

> The dystopia might arrive someday, but not now, and not in the near future. Democrats (yes, it was all democrats) got it done seven years ago.

Fortunately most people (or someone they care about) are likely enough to suffer from a genetically-linked, costly health issue that it seems unlikely that discrimination would find much support.

Strange, then, that the Republican-dominated House just passed a bill weakening protections against discrimination on the basis of genetic information.

https://www.theatlantic.com/health/archive/2017/03/genetic-d...

I think this is just due to a misunderstanding of how genetic traits and risks work. Insurance companies may be angling for the government to be the insurer of last resort for the most costly genotypes.

If there is anything that it makes sense to utilize a risk pool for, it is the random chance associated with heredity. Aside from a small number of well known, highly hereditary genetically-linked conditions, most people just reproduce without thinking much about the probability that their offspring will end up with an unlucky genotype. This is mainly because there is so much chance involved and a low probability of the child getting a "costly" genome, at least the cost impact of drawing a 99% probability genome is far less than the impact of an additional year of maternal age, etc. Even for the most strongly hereditary diseases, random chance still dominates the impact of heredity on the chance of the offspring getting the disease.

Also, evolution ensures that most "harmful" genes are also helpful in some way... so while there is likely an expected dollar cost of a given allele, I don't think there's enough variation in overall dollar cost to make it worth the hassle for anyone (including insurers), which points again to the notion that they are simply looking for corporate welfare.

For comparison, post 9/11 the insurance industry extracted a big handout when the US Government became the insurer of last resort for terrorism related claims > $1B. My take is that genomes that are highly likely to cost more than a standard deviation greater than average are rare enough that it wouldn't even make sense from a competitive perspective for insurance companies to try to evade issuing policies to such people.

I believe it is central to progress in health care in the U.S. that discrimination on the basis of human genetic composition (including resident microfauna/microflora) be fundamentally outlawed, with personally attachable criminal liability for breaching said law.

We could be learning a lot from population studies, but said population needs to be assured that participation -- willing or otherwise -- will not result in a personal negative outcome. Such as being denied insurance, employment, credit (yes, you bet financial institutions will use such data if they can), etc.

Those would be the real "death lists". Unlike the fear-mongering we've seen to date about e.g. end-of-life discussions. (Having just spent time and caregiving with an old friend going through their own end-of-life experience, I feel fairly confident in saying that most people would prefer to have those conversations and an understanding with family and friends as to what they do and don't want. This person was suffering at the end and basically pleading with her deity to take her and end that suffering. We made her as comfortable as possible; nonetheless, she didn't want that condition to continue -- e.g. through exceptional medical intervention -- longer than necessary.)

If you want my genetic make-up, guarantee me it is not going to be used against me in terms of equal participation in society. I didn't "choose" it, and I'm doing the best I can to work with and take care of what I have. Help me do that, instead, on an equal footing with my fellow citizens.

Note that I am not saying, 'Guarantee me equal outcomes.' I am saying guarantee me equal access to participation, including and especially in health care.

Also, don't pattern-match me against every piece of crime evidence, when such testing has repeatedly been shown to have limits to its accuracy and the courts shown to have no understanding nor accommodation for such inaccuracies nor the very real limits of the technology and attendant crime scene and evidence management.

Could CRISPR be used to fix this?

Exactly what I was thinking... but when will you (and society) be comfortable with a therapy that goes into your brain cells and edits their DNA?

I'm ready

>but when will you (and society) be comfortable with a therapy that goes into your brain cells and edits their DNA?

Seconds after some early adopters have shown that it works and has no side-effects?

No on will disagree with that answer.

Lots of people will argue vehemently about what it takes to be certain that there are no side effects.

:) I'll bite. "Seconds" is not nearly long enough for me to be confident there are no adverse effects. Deleterious effects from monkeying with genes can take years or decades to show up (via increased risk to certain cancers, etc.)

Note that I didn't say: "Seconds after some early adopters tried it", but "Seconds after some early adopters have shown that it works and has no side-effects".

Which could be a few decades after some early adopters tried it -- e.g. until they died in their 80s or 90s with perfect brain function...

So, seconds after that...

Right, that was my understanding.

I suppose my point was that kornork seemed to me to be talking about safety and in that context, saying you'd do it after it is shown to be safe isn't saying a lot.

Aha, I took the parent's comments to be about the ethical and/or existential questions.

After all we do various kinds of brain altering operations and take brain altering drugs. DNA is not our personality, just how our body is built.

>No on will disagree with that answer.

I disagree; there'll probably be a bunch of people who will vehemently disagree with "artificially" altering humans like this, even just to correct a genetic defect. They'll say it's "unnatural", or that it "isn't God's will", or somesuch.

I don't know that they'll be a big portion of the population, but they'll be out there.

When the alternative is your mind falling apart, I'd expect quite a few people will risk the side effects. I certainly would.

Not an expert on the technique, but if it's possible to "edit" a gene, how would the altered gene be propagated to billions of neurons (assuming that's where the gene has effect)? The idea of modifying the gene in situ is a technical issue that would seem awfully hard to resolve.

Of course the whole idea is predicated on having sufficient knowledge of the gene's effect. The reported finding awaits replication, and quite likely further refinement before practical applications are feasible.

I would assume editing In Situ would require a retrovirus or something similar. Seems incredibly risk, but then again, so was open-heart surgery until we figured out all the things that will go wrong.

There are more cells in a body than stars in the universe; I'm under the impression that the first CRISPR therapies will be on early-stage embryos where the number of cells is still manageable and we can feasibly edit all of them. Would be happy to hear more from someone who knows more about the latest in gene therapy though, my training here is getting a little dated.

1 x 10^24 stars vs 3.2 x 10^13 cells.

Estimates from a quick google.

Darn, internet info foiled by more internet info...

This kind of point mutation is exactly the kind of issue that CRISPR is capable of fixing. This can be done today. The issues are specificity and delivery of the fix. Delivery is hard to do non-invasively without side effects, and the 'fix' must not only correct only the error, but also not 'correct' a non-error elsewhere. Getting the 'patch' to the right place, and making sure the patch only affects exactly this sequence is now the hard part - actually creating the patch is doable today.

Would anyone be kind enough to post instructions for finding this in your 23andme data?

Reply on news.ycombinator.com