New study finds vast majority of diseases have only a very small genetic contribution of 5% to 10% at most (so much for 23andme testing)

[Hip]

A new large meta-analysis has shown that for the vast majority of diseases, genes only play a very minor role in determining whether you get the disease or not. For the majority diseases, this study found the genetic contribution to disease development is only 5% to 10% at most.

Professor David Wishart, one of the study authors, says:

DNA is not your destiny, and SNPs are duds for disease prediction

And he goes on:

It is becoming increasingly clear that the risks for getting most diseases arise from your metabolism, your environment, your lifestyle or your exposure to various kinds of nutrients, chemicals, bacteria or viruses

So much for SNP testing services like www.23andme.com.


Though the study did identify a few diseases which have a higher genetic contribution, including Crohn's disease, celiac disease and macular degeneration, which have a genetic contribution of about 40% to 50%.

But in general, this study shows that genes and SNP mutations are not the answer when trying to understand how disease arises.

Myself, I subscribe to Professor Paul W. Ewald's views: he has always argued that genes are not key factors when seeking out the causes of disease, and he thinks most chronic diseases will turn out to be caused by infectious pathogens (viruses, bacteria, etc). Pathogens are environmental factors that many researchers and the general public overlook, but these may be playing a major role in triggering and maintaining disease.



Article about this new paper here.

 

[Hipsman]

For the majority diseases, this study found the genetic contribution to disease development is only 5% to 10% at most.

This is very interesting, I was just looking at DRACO longecity thread yesterday, and found your post about genetic contribution very informative, I think others here will find it interesting too.

 

[Hip]

This is very interesting, I was just looking at DRACO longecity thread yesterday, and found your post about genetic contribution very informative, I think others here will find it interesting too.

The concordance rates stuff? Yes, it's definitely an interesting concept, I'll post it here:



Concordance Rates — How To Tell If a Disease Is Caused by Genetics or Environmental Factors

How do you work out whether a disease is mainly genetic or mainly environmental? That can be achieved by looking at concordance rates: where you look at identical twins, and when one twin develops a disease, you observe if the other twin develops it also.

For a purely genetic disease, if one identical twin gets the disease, then the second twin will always get it too. But if the disease etiology involves environmental factors, then the second twin will often not get the disease. The chances that the second twin gets the disease is the concordance rate for that particular disease.

A 100% concordance rate means the disease is purely genetic, and 0% concordance rate means the disease is purely environmental. Percentages in between mean the disease is a combination of genetic and environmental.

With lots diseases, the concordance rate is low, which shows that the disease is primarily environmental, and not genetic. So we know for lots of diseases, environmental factors such as toxins and pathogens must play a key role.

For example, the concordance rate for type 1 diabetes is just 34%, which suggests T1D is mainly caused by environmental factors. Ref: here.

Type 1 diabetes has long been associated with a common virus called coxsackievirus B4. This virus can infect and destroy the insulin-producing cells of the pancreas. See this paper for example. So there is an easy to understand mechanism by which coxsackievirus B4 might cause diabetes.

 

[Hipsman]

The concordance rates stuff?

yes, I meant this.

 

[Wishful]

By that, we should be thinking of our predisposition to ME as being due to environmental factors rather than genetic ones. That would fit with ME being more common in today's more challenging (toxins, diet, stress, etc) environment, assuming that ME actually is more common today, rather than just being more recognized.

Has anyone mapped out the prevalence of ME today? Difficult, since many areas might not even recognize it as a disease, but I suppose there are mathematical tools for dealing with that. It would be interesting if it showed that ME is more common in people who grew up in an urban environment, or who consumed lots of sugar...or didn't eat their vegetables.

 

[Rufous McKinney]

or didn't eat their vegetables.

I definately did not eat my vegetables. (all canned, all horrible)

 

[ebethc]

tl;dr ... averages suck b/c they're not that meaningful or actionable
-
Re CFS: whatever is wrong w my health is definitely genetic....
I'm adopted and was raised by parents who were extremely health conscious... My adoptive mother and her family are VERY intense exercisers (marathoners, etc) and diet means "calories in vs calories out" w plenty of veggies, and what most americans would consider a "sensible diet"... She's 5'5" 100 lbs and has never counted a carb in her life or bought organic anything... She has a lot of health problems in her family that I don't have, and eating low calorie diet w a ton of cardio exercise keeps her healthy...eg She's avoided the cancer that 3 of her siblings have had... (2 recovered, 1 didn't. 1 of the siblings who recovered is a health nut, too, but the other one only plays golf for "exercise"..The one who died was the only sibling in the family who eats bad and doesn't exercise..).. She def raised us to get nutrients from food vs pills/supplements. She believes that good health is discipline and sensible choices (yes, but sometimes NOT).

I met my biological mother in college, and she + her oldest daughter have a symptom overlap w me of ~90%. they live in a different part of the country than me, and she raised her family all over the world due to her husband's profession (ie, how much is really environmental??). She seems health conscious, but I didn't grow up w her so I can't say to what degree (ie, my adoptive mother is OBSESSED w health)... So, lifestyle and choices can't be blamed for everything.

With that said, I could be doing something "wrong" every day because I don't understand what the problem is.. I wouldn't be on this board if I didn't think there's something to understand and change. Taking care of yourself is extremely important, but what does "taking care of yourself" even mean if you're the canary in the coal mine? We're all on PR to share our n=1 experiments, trying to find out.

Re Huntington's Disease (as an alternative example):
I have a friend whose wife has Huntington's disease, which is VERY much a genetic disease... He has two kids and is worried sick they have the gene... He hasn't had them tested b/c it's their choice when they're old enough to decide. He works tirelessly on fundraising to find a genetic-level cure..
https://www.hda.org.uk/huntingtons-... your parents is,75 % chance of inheriting it.

Re Genetics, 23andme, etc.: I think it's the COMBO of genes that screws you over....
I'm surprised ppl don't acknowledge this or see it more... and I wonder if AI will help us understand this better.. e.g., ppl look up one gene in 23andme and don't necessarily understand the whole pathway... I think our bodies are incredibly adaptable and if something's broken along some pathway, we can adapt to accommodate the problem, BUT, if you have the wrong COMBO of bad genes, your body can't overcome being repeatedly thwarted and breaks down .. so looking at a discrete gene can really send you down a rabbit hole...

 

[Learner1]

@Hip

So much for 23andme? I beg to differ... I have:

• hemochromatosis caused by HFE genes
• Factor 2, caused by F2 genes
• KIT genes predisposing me to MCAS
• 14 thyroid genes with about 200 mutations and I have thyroid issues
• Genes that lead to low folate, low B12, low tetrahydrobiopterin, low MnSOD, and low vitamin D

My sister died of a condition caused by a rare RASA1 mutation.

There are people with mitochondrial diseases, Tay Sachs disease, Parkinson's disease, breast cancer and many other diseases caused by genes.

I'm surprised ppl don't acknowledge this or see it more... and I wonder if AI will help us understand this better.. e.g., ppl look up one gene in 23andme and don't necessarily understand the whole pathway... I think our bodies are incredibly adaptable and if something's broken along some pathway, we can adapt to accommodate the problem, BUT, if you have the wrong COMBO of bad genes, your body can't overcome being repeatedly thwarted and breaks down .. so looking at a discrete gene can really send you down a rabbit hole...

This is very true.

 

[Hip]

By that, we should be thinking of our predisposition to ME as being due to environmental factors rather than genetic ones.

According to this study, the monozygotic (identical) twins proband concordance rate in ME/CFS is 55%, whereas the dizygotic (fraternal) twins rate is 19%.

When the identical twin concordance rate is substantially higher than the fraternal twin concordance rate, that suggests there are some genetic causal factors which contribute to the development of the disease.

 

[Hip]

So much for 23andme? I beg to differ... I have:

Everybody has some problematic mutations, and yes there are some purely genetic diseases; but that's not really what Wishart's research is getting at.

Let me explain the scientific background context and history:

The human genome was only fully sequenced quite recently, in 2003. I actually remember President Clinton announcing the completion of this project.

The human genome project took an enormous effort and was costly, but at the time it was believed that faulty genes were probably the cause of most diseases, so scientists considered it worth all the effort.

However, unfortunately it has turned out that in general, the idea of faulty genes causing common diseases was just wrong. We have now learnt that faulty genes rarely cause disease; at best faulty genes might mildly predispose us to a disease. But they certainly are not a viable explanation for why common diseases arise.

This old Guardian article from 2011 explains the human genome project fiasco:

Among all the genetic findings for common illnesses, such as heart disease, cancer and mental illnesses, only a handful are of genuine significance for human health. Faulty genes rarely cause, or even mildly predispose us, to disease, and as a consequence the science of human genetics is in deep crisis.

So looking at our genes in order to find the cause of common diseases has been a disaster.

The Guardian article goes on:

The failure to find meaningful inherited genetic predispositions is likely to become the most profound crisis that science has faced. Not only has the most expensive scientific project ever conceived failed to reach a goal it assured the world it would achieve, but there is also the ticklish problem of why the headlines have been so consistently discrepant with reality. As the failures to find significant genes have accumulated, geneticists have remained silent.

There are still important decisions to be made. The Collins paper proposed a no doubt expensive and open-ended search among hitherto disregarded genetic locations. We should be under no illusions, however. The likelihood that further searching might rescue the day appears slim. A much better use of that money would be to ask: if inherited genes are not to blame for our commonest illnesses, can we find out what is?

So the Guardian points out that since genes are not the answer to what causes common diseases, what science must now do is figure out what does cause these diseases.

Wishart's work only underscores what we already know: that the answer is not to be found in the genes. Wishart says that:

It is becoming increasingly clear that the risks for getting most diseases arise from your metabolism, your environment, your lifestyle or your exposure to various kinds of nutrients, chemicals, bacteria or viruses

So we need to start looking at these environmental factors to explain disease, environmental factors like chronic pathogenic infections and toxic exposures.

 

[ebethc]

According to this study, the monozygotic (identical) twins proband concordance rate in ME/CFS is 55%, whereas the dizygotic (fraternal) twins rate is 19%.

When the identical twin concordance rate is substantially higher than the fraternal twins concordance rate, then that shows there are some genetic causal factors which contribute to the development of the disease.

the disease.

that's assuming that CFS is ONE illness... ("the" disease)... I think part of the problem in finding answers is that it's multiple illnesses, not one illness w one root cause and multiple tributaries... symptom based diagnosis sucks

info re twins is interesting

 

[ebethc]

So we need to start looking at these environmental factors to explain disease, environmental factors like chronic pathogenic infections and toxic exposures.

I agree that the expectations were too high for the genome mapping... one person explained it to me in the 90s as, We will have a blueprint. It's like you've been tinkering around w a car engine, hoping to fix it, then someone hands you the manual and specifications so you finally know what to do! <very breathless>

now the same thing is going on w microbiome research... it's going to give us so much insight, etc. I'm very interested to see where it goes, and I'm def bought into it... However, do you think the same thing is happening all over again? too much expectation, too much single-mindedness?

 

[Hip]

now the same thing is going on w microbiome research... it's going to give us so much insight, etc.

My guess is that the gut microbiome research will go the way of the human genome project: plenty of hype and expectation — and the gut microbiome is currently in vogue in science, so easier to get research grants — but I think the final results will probably be lackluster. This article says: "microbiomics risks being drowned in a tsunami of its own hype".

The gut microbiome I think may turn out to play some kind of role in illness, but I suspect that its role will be a minor one.

By contrast, the microbiome in the body as a whole, which would include viruses that infect various organs, I suspect will turn out to be the main causal factor in many everyday chronic diseases. This is the view that Professor Paul Ewald is always trying to convey.

 

[Learner1]

Wishart's work only underscores what we already know: that the answer is not to be found in the genes. Wishart says that:
So we need to start looking at these environmental factors to explain disease, environmental factors like chronic pathogenic infections and toxic exposures

It is both genetics and environmental factors. In Parkinson's, there are known genes, as well as factors like toxins and the microbiome. Same with Alzheimer's.

In mitochondrial diseases and hemochromatosis, it is the gene, yet how much the gene is expressed varies form patient to patient.

However, in many cases, just because someone has a bad gene, it doesn't necessarily doom them to tragedy. They have a tendency, but they may have other genes counteracting the bad effect, or they can manipulate their local environment to budge the gene to better behavior.

I've found it extremely empoweing to be able to have and use my genetic information from both 23andme and a doctor ordered WES and mtDNA test. It's been useful to understand the genes affecting my health problems, to be able to match treatments to the situation, and to be able yo improve my health. Yes, environmental factors have played a part, but the effect of some genes can be pretty powerful.

The gut microbiome I think may turn out to play some kind of role in illness, but I suspect that its role will be a minor one.

Week after week, more medical journal articles come out linking the gut microbiome to certain diseases. A few recent examples:

https://www.sciencemag.org/news/202...bacteria-may-alter-how-you-think-feel-and-act

https://neurosciencenews.com/als-microbiome-16393/

https://www.nature.com/articles/s41598-020-62224-3

https://www.sciencealert.com/scient...nking-gut-bacteria-with-neurovascular-disease

https://www.michaeljfox.org/news/ask-phd-parkinsons-and-microbiome

https://www.nih.gov/news-events/new...d-brain-blood-vessel-abnormality-gut-bacteria

https://www.pnas.org/content/early/2020/05/28/1922498117

 

[LINE]

Has anyone talked about epigentics? (that is how the environment changes the genetic expressions)

A family member is a medical researcher in infertility. He is pretty conservative, also published in medical journals. I asked him what the latest findings were on infertility and he surprised me by saying that epigenetics seems the most viable - he went on to describe that it seems the environmental factors were the causative factors in epigenetic shifting.

Another family member works at the FDA and she told me the scientists there said there is a big problem in chemicals in the environment, more specifically that combinations of chemicals. For instance, they know about window cleaners, but what if you spray a window cleaner on plastic, it forms a new compound. They have no idea what these do inside the body.

One of my treatment protocols involved detoxification which IMO has helped.

 

[Snow Leopard]

BTW, The study was published in December.

https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0220215

Using these data we found a small number of GWA studies with SNP-derived risk predictors that have very high AUROCs (>0.75). On the other hand, the average GWA study produces a multi-SNP risk predictor with an AUROC of 0.55. Detailed AUROC comparisons indicate that most SNP-derived risk predictions are not as good as clinically based disease risk predictors.

 

[Hip]

Has anyone talked about epigentics? (that is how the environment changes the genetic expressions)

A family member is a medical researcher in infertility. He is pretty conservative, also published in medical journals. I asked him what the latest findings were on infertility and he surprised me by saying that epigenetics seems the most viable - he went on to describe that it seems the environmental factors were the causative factors in epigenetic shifting.

Epigenetics is an important line of investigation, but unfortunately still in its infancy compared to genetics, and studying epigenetics is complicated by the fact that there is more than one mechanism at play (the body has several mechanisms by which it controls gene expression).

But it makes sense that epigenetic factors might play an important role in disease, as epigenetics determines whether entire genes are expressed or not (whether the gene is turned on or off), and the degree to which a gene is expressed.

Whereas when we consider genetics's role in disease, remember that all genes are identical from one human to the next, apart from some very small areas of the gene, where you get these polymorphism (mutation) variations from one person to the next.

So everyone's genes are largely the same, except for the polymorphisms, which we call SNPs (single nucleotide polymorphisms). It was hoped that these SNPs would be the basis of common chronic diseases, but this has turned out not to be the case.

Furthermore, Professor David Wishart's very large meta-analysis concludes SNPs are duds for disease prediction.

 

[Learner1]

Furthermore, Professor David Wishart's very large meta-analysis concludes SNPs are duds for disease prediction.

Let's just take one disease, hemachromatosis:

• rs1800562 in the HFE gene, also known as C282Y (risk genotype AA) can cause a serious form of Hemochromatosis, and accounts for about 85% of Hemochromatosis patients. About 1 in 200 people of European ancestry is a carrier of this mutation.
• rs1799945 in the HFE gene, also known as H63D (risk genotype GG), can cause a mild form of Hemochromatosis.

Those who are carriers (one risk allele) of both the C282Y and H63D variants may be affected by a mild form of Hereditary Hemochromatosis, but only if they are on separate chromosomes (i.e. one is on the maternally inherited copy of the HFE gene, and the other is on the paternally inherited copy). Note that microarray based genotyping (such as is done by 23andMe or FamilyTreeDNA) cannot determine if the variants are on the same allele (i.e. chromosome).

Other SNPs may influence the risk of serious hemochromatosis:

• rs235756 in the BMP2 gene may lead to higher risk of developing hemochromatosis in carriers of two A alleles of rs1800562.
• rs2280673 in the RAB6B gene is associated with serum transferrin levels. The risk allele appears to be C.
• rs1800730, called i3002468 by 23andMe, in the HFE gene, also known as S65C (risk allele T), may be associated with a third Hereditary Hemochromatosis variant, but more research is needed.

There are several types of hereditary hemochromatosis other than the classic (Type 1) form primarily being discussed in the preceding paragraphs. Type 2 hemochromatosis is the result of mutations in the HJV or the HAMP genes, while TFR2 gene mutations cause Type 3 hemochromatosis; like Type 1, Type 2 and Type 3 are recessively inherited conditions. In contrast, dominant mutations in the SLC40A1 gene can lead to Type 4 hemochromatosis.

 

[Hip]

Let's just take one disease, hemachromatosis:

Hemochromatosis is a hereditary genetic disease, so obviously it's going to be caused by SNP mutations. Nobody is disputing that. Professor Wishart's study is not suggesting purely genetic diseases are not caused by SNPs, as we know they are.

Wishart's study examined a wide range of hundreds of common chronic diseases that afflict humanity, and discovered that SNPs do not predict who gets these diseases. So the genetic contribution to most diseases appears minimal.



Wishart's study results are provided in this database: https://gwasrocs.ca/study_simulations

For each disease, he calculated the heritability factor using genome-wide association (GWA). If you scroll down the list of diseases, you see that in most cases, the heritability is very low, less than 10% (ie, less than a factor of 0.1 ). So for most diseases, it appears genes are playing only a minimal role in disease causality.

Exceptions include celiac disease, which this database shows has a heritability of 58.2% (ie, a heritability factor of 0.582). So for celiac disease, genes play an important role in who gets the disease and who does not.

Unfortunately ME/CFS is not in Wishart's database.

 

[Hip]

I've taken some diseases from Wishart's database and posted their heritability value below. A heritability of 0% means that no genetic factors are involved in disease causality, and a heritability of 100% means the disease is purely caused by genetic factors.

Here are the heritability values calculated by Wishart's study for a few diseases:

Celiac disease 58.2%
Systemic lupus erythematosus 14.8%
Alzheimer's disease 10.6%

Sjogren's syndrome 9.6%
Parkinson's disease 4.3%
Major depressive disorder 4.3%

Chronic hepatitis B infection 3.7%
Rheumatoid arthritis 3.4%
Type 2 diabetes 3.3%

Type 1 diabetes 3.2%
Obesity 2.3%

Note that each disease appears multiple times in the database, and on each appearance the heritability has a different value. I don't fully understand it, but each appearance relates to a specific study. The heritability values I posted just above are the highest values in the database for the disease.