Hello There, Guest!  

Epigenetics [split from Ethnic Differences in the UK]

#1
(2015-Nov-20, 18:04:02)Chuck Wrote: What does it matter?


It matters a lot since Humans have the same genes. Its only the coding thats different, a very small amount of the coding. What epigenetics has proven is that this small amount of coding is changeable via environment and then some.

The sequence of the sequence does not get affected(as far as I know) but its chemical make up does, so does the biological make up outside of the sequence and even the RNA which is the messenger between the sequence and gene. These are also most importantly heritable.

In lab experiments on animals, the effects have been pretty large eg: Agouti mouse study. Its no joke, genetically identical mice looked like different species, even though the detectable epigenetic effect(methylation) was small. Some of the progeny actually had the phenotype effect even without detectable methylation.

It will make a big difference to the argument, to selection theories, to everything and I highly doubt its in your favour.

Also don't bother talking to me about the validity of your race concept. Its not of any concern to me.
 Reply
#2
(2015-Dec-14, 08:47:50)Zoidberg Wrote:
(2015-Nov-20, 18:04:02)Chuck Wrote: What does it matter?


It matters a lot since Humans have the same genes. Its only the coding thats different, a very small amount of the coding. What epigenetics has proven is that this small amount of coding is changeable via environment and then some.

The sequence of the sequence does not get affected(as far as I know) but its chemical make up does, so does the biological make up outside of the sequence and even the RNA which is the messenger between the sequence and gene. These are also most importantly heritable.

In lab experiments on animals, the effects have been pretty large eg: Agouti mouse study. Its no joke, genetically identical mice looked like different species, even though the detectable epigenetic effect(methylation) was small. Some of the progeny actually had the phenotype effect even without detectable methylation.

It will make a big difference to the argument, to selection theories, to everything and I highly doubt its in your favour.

Also don't bother talking to me about the validity of your race concept. Its not of any concern to me.


Environmental insults can also change the DNA sequence, i.e. some environmental insults cause mutations (e.g. carcinogens).

Is there any meta-analysis of these lab experiments on animals yet? Publication bias is an obvious issue.

Look what happened to the SES x heritability idea. https://twitter.com/KirkegaardEmil/statu...1466040320 Small outlier study gets most of the attention. A familiar pattern.

Do you remember what happened to e.g. stereotype threat and Pygmalion effect? Mostly or entirely publication bias.

Why should we expect intergenerational epigenetic phenomena to be anything else? History teaches us to be skeptical about these kinds of claims since they appear to have a low base rate of panning out.

Recall that cognitive ability is controlled by 1000s of genes. One would need to do a lot of methylation changing to alter that I suspect. This is in contrast with the few-gene traits favored by animal experimentalists.
 Reply
#3
(2015-Dec-14, 09:27:55)Emil Wrote: Is there any meta-analysis of these lab experiments on animals yet? Publication bias is an obvious issue.

Look what happened to the SES x heritability idea. https://twitter.com/KirkegaardEmil/statu...1466040320 Small outlier study gets most of the attention. A familiar pattern.

Do you remember what happened to e.g. stereotype threat and Pygmalion effect? Mostly or entirely publication bias.

Why should we expect intergenerational epigenetic phenomena to be anything else? History teaches us to be skeptical about these kinds of claims since they appear to have a low base rate of panning out.

Recall that cognitive ability is controlled by 1000s of genes. One would need to do a lot of methylation changing to alter that I suspect. This is in contrast with the few-gene traits favored by animal experimentalists.


But transgenerational epigenetic inheritance is a proven biological fact, there is no need for scepticism or meta analysis.
http://www.pnas.org/content/112/7/2133.abstract
http://www.sciencemag.org/content/345/6203/1515
http://stke.sciencemag.org/content/7/344/ec263.abstract

The agouti mouse experiment was a softer study when it came to the mark(methylation) but even that actually measured the biological changes from gene expression to phenotype. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2822875/

Also, as you can see its not only "single gene" effects. The agouti mosue model found multiple effects from adult-onset obesity, diabetes and tumorigenesis on top of the hair color effect.

They have studies on behaviour too.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3217197/

Also cytosine methylation(which I assume is what you are thinking of) is not the only epigenetic mechanism, there is much more going on outside of the sequence, Eg: RNA, and histone modification. On top of that there are massive amounts of methylation in the human body especially in the brain.
http://www.sciencemag.org/content/348/6239/1094

Like I said, its no joke. There is a reason why people like Burt and Simons are so confident in their assertions.
 Reply
#4
I see a few worm studies, and some mouse studies with hidden sample sizes. They don't mention them outside, you have to look for them. For instance, Paternal Transmission of Stressed-Induced Pathologies does not mention sample size in abstract or methods section. However, one can get it from their tests statistics "(Figure 1A, n=8-9 per group (t=2.229 df=15; *p<0.05).". I don't know what the asterisks are supposed to mean.

I know of one human study with tiny sample. http://www.sciencedirect.com/science/art...2315006526

I don't see any meta-analyses.

The evidence you posted is very weak if the desired conclusion is something like "Heritability estimates from familial behavioral genetic studies are totally untrustworthy.".

PS. Why do you post epigenetics stuff in a thread about ethnic differences in the UK? Do I need to split the threads?
 Reply
#5
(2015-Dec-15, 11:37:43)Emil Wrote: I see a few worm studies, and some mouse studies with hidden sample sizes. They don't mention them outside, you have to look for them. For instance, Paternal Transmission of Stressed-Induced Pathologies does not mention sample size in abstract or methods section. However, one can get it from their tests statistics "(Figure 1A, n=8-9 per group (t=2.229 df=15; *p<0.05).". I don't know what the asterisks are supposed to mean.

I know of one human study with tiny sample. http://www.sciencedirect.com/science/art...2315006526

I don't see any meta-analyses.

The evidence you posted is very weak if the desired conclusion is something like "Heritability estimates from familial behavioral genetic studies are totally untrustworthy.".

PS. Why do you post epigenetics stuff in a thread about ethnic differences in the UK? Do I need to split the threads?


I apologize, but could you split it? I'll continue there in the other post.
 Reply
#6
(2015-Dec-15, 11:37:43)Emil Wrote: I see a few worm studies, and some mouse studies with hidden sample sizes. They don't mention them outside, you have to look for them. For instance, Paternal Transmission of Stressed-Induced Pathologies does not mention sample size in abstract or methods section. However, one can get it from their tests statistics "(Figure 1A, n=8-9 per group (t=2.229 df=15; *p<0.05).". I don't know what the asterisks are supposed to mean.

I know of one human study with tiny sample. http://www.sciencedirect.com/science/art...2315006526

I don't see any meta-analyses.

The evidence you posted is very weak if the desired conclusion is something like "Heritability estimates from familial behavioral genetic studies are totally untrustworthy.".


See the problem here is that you are thinking of these studies like they are heritability studies. Some of them like the one you posted are, but the ones I linked aren't, as they actually measure the biological change in the gene, the phenotype and also the epigenetic marks(as with the worm and nemadote studies) being past on. Sample size does not matter for such things since the fundamental biology of the gene is the same for all of us. We share a lot of our genes with these organisms, even plants. With mice we share about ~88% of our genes and with animals like roundworm we share ~38% or so. We share ~73% with zebra fish! Even the genes that are not shared still work in pretty much the same fundamental way.

Either way, the studies I provided outright prove transgenerational epigenetics(especially the worm studies). They showed the induced mark being transferred visually, its undeniable. You must accept that in the least.

The agouti mouse study used genetically identical mice, they automatically prove that the same genes can can cause massive differences regardless of the sequence variation(which is everything to your heritability model):
[Image: mice.jpg]

The big issue here is that everyone has the same gene, IE the basic hardware, the thing making the protein and expressing it. The difference is mostly one letter in the coding sequence and even that little bit can be changed by environment and the biological mechanisms that mediate these changes are extremely common within the human and animal physiology, especially within the brain.

There is no way in hell that the hereditarian model is unaffected by this.

Now, I am not fully siding with Burt and Simons yet, but what they said is true from what I have found so far and devastatingly so.
 Reply
#7
Zoidberg Wrote:See the problem here is that you are thinking of these studies like they are heritability studies. Some of them like the one you posted are, but the ones I linked aren't, as they actually measure the biological change in the gene, the phenotype and also the epigenetic marks(as with the worm and nemadote studies) being past on. Sample size does not matter for such things since the fundamental biology of the gene is the same for all of us. We share a lot of our genes with these organisms, even plants. With mice we share about ~88% of our genes and with animals like roundworm we share ~38% or so. We share ~73% with zebra fish! Even the genes that are not shared still work in pretty much the same fundamental way.


Sample size matters because there is always some within group variation in measurement (if only because of measurement error). This means that group means have some uncertainty in their estimates, which is larger when samples are small. The studies you posted acknowledge this indirectly because they use inferential statistics. If there was no within group variation (and I don't mean just by chance or with N=1), inferential statistics would not be necessary because group means would have no estimation error. One could just treat the sample means as population means.

Zoidberg Wrote:Either way, the studies I provided outright prove transgenerational epigenetics(especially the worm studies). They showed the induced mark being transferred visually, its undeniable. You must accept that in the least.


I don't have problems with trans-generational epigenetics itself, only with some claims some people think are supported by this phenomenon.

Zoidberg Wrote:The agouti mouse study used genetically identical mice, they automatically prove that the same genes can can cause massive differences regardless of the sequence variation(which is everything to your heritability model):


Careful with the strawmen. I have not stated that transgenerational epigenetics plays no part in all traits.

Also note that genetically identical mouse are not in fact entirely so, just like human MZ twins.

I've not seen any reputable expert claiming (in recent decades!) that DNA variation is responsible for all observed differences in cell behavior. We know this to be false because epigenetics is responsible for cell differentiation within organisms. E.g. the cells in my foot and my brain have almost the same DNA sequence (de novo mutations aside). That is, unless I'm a chimera. But we know that cells in those two places behave very differently, due mostly (as I understand, someone correct me if I'm wrong) to epigenetics.

Zoidberg Wrote:The big issue here is that everyone has the same gene, IE the basic hardware, the thing making the protein and expressing it. The difference is mostly one letter in the coding sequence and even that little bit can be changed by environment and the biological mechanisms that mediate these changes are extremely common within the human and animal physiology, especially within the brain.


Unclear. Everyone has the same gene(s)? Does everyone refer to the mice in the study or humans? Humans do not have all the same genes, some are recent adaptions since races split. E.g. lactose tolerance, malarial resistance or height adaptions.

But humans mostly have the same genes, yes. However, there is variation (as you know), some of which is functional, most of which is probably not.

Zoidberg Wrote:There is no way in hell that the hereditarian model is unaffected by this.


What is the thing you call "the hereditarian model"? Hereditarian is derived from hereditary, so something that runs in families. Typically, it is used to mean genetic effects. One can talk about a hereditarian model of differences within a population (within group heritability), and one can talk about a hereditarian model of differences between populations (between group heritability). Such models do not imply any strict DNA-only claim, i.e. they are consistent with epigenetic effects, both within organisms and across generations (transgenerational epigenetics).

It is possible, for instance, that of the observed group differences in height across all human populations, 25% of these are due to nutritional variation, 37.5% due to epigenetic variation and 37.5% due to genetic variation. It is also possible that if we look at some other set of populations, say, Europeans, the observed differences in height are nearly 100% due to genetic variation.
 Reply
#8
Zoidberg I am on your side. I have read all the recent epigenetics research and I convinced that many individual differences are mediated by epigenetic inheritance of ancestral environments. That racial differences can be explained by epigenetic corruption of the genome of blacks as a result of colonization and slavery due to stress, poverty, trauma, malnutrition and lack of societal development.
 Reply
#9
The control of epigenetic change is coded in DNA too. It doesn't code amino acids and is referred by some as the "dark matter" in our DNA. It is evolved. Animals evolved such regulatory function and the function must have some fitness value. It is unlikely that human evolved such regulatory function to suppress its offspring’s' brain because such DNA are harmful to the genome and cannot be passed until today. Although it isn't surprising if there is any coincidence. (i.e. epigenetic changes that serve other functions but suppress mental ability.) However its magnitude is likely to be small because intelligence is perhaps the most important human trait and suppressing it makes no contribution to an individual's survival and pass on its genes.
 Reply
#10
(2016-Jan-17, 05:21:39)General-Factor analyst Wrote: Zoidberg I am on your side. I have read all the recent epigenetics research and I convinced that many individual differences are mediated by epigenetic inheritance of ancestral environments. That racial differences can be explained by epigenetic corruption of the genome of blacks as a result of colonization and slavery due to stress, poverty, trauma, malnutrition and lack of societal development.

Got it. What a pity your ancestors, regardless of race, must have been through the worst colonization and slavery. They must have suffered stress, poverty, trauma, malnutrition and lack of societal development worse than anyone else ever have.
 Reply
 
 
Forum Jump:

Users browsing this thread: 1 Guest(s)