> socioeconomic status (SES) is a powerful determinant of human health and disease, and social inequality is a ubiquitous stressor for human populations globally
It is common to see why this could be a big problem for poor people as they have to worry about health, food, etc for day to day living while we take them as granted and can focus on other things like writing code, going to meetings, and the like without worrying about where the next meal is going to come from!
My observations is that our society puts too much emphasis on having people prove they are self-reliant, instead of improving its population's productivity. In this dichotomy, the former is detrimental to the society, the latter would enhance it.
I think focusing on productivity would go further than the dichotomy's of "welfare or not", where people are mostly skeptical of whether a subsidized poor person is contributing to society with taxpayer's money.
Relegating the role of governance to productivity allows even prisoner rehabilitation to change. It allows jailable offenses to be viewed under the lens of whether this is useful to the productivity of society, instead of simply punishing someone.
I agree with the sentiment of changing the narrative, but productivity seems to already be a big part of western society and it seems pretty soul-less. Why not focus on well-being of everyone?
I'm not sure what fraction of paid jobs even contribute greater than their cost to society. The War on Poverty has spent trillions, and poverty has increased. A lot of industries obviously don't contribute (tax attorneys, lobbyists, bureaucrats, administrators (see the Rise of the Administrator in healthcare), even much of law enforcement, academia, or regulators). Actually even my first high-paid job working for a financial tech company, I'm convinced didn't contribute to society. Most of our work was mandated by regulatory laws, and most of our customers were capture by monopolistic competition or downright aggressive, questionable sales.
I don't see anyone saying this, but I think fixing the bottom 10% of society requires fixing the top 90%. I see no way that happens. Nobody accepts blame for being wrong, and most people's cognitive capabilities are filled by an evening watching cable TV.
The bottom with disruptions to their income are detrimental to society’s health. Public sector monetary solutions are untolerated and private sector solutions of employment are often inaccessible or inadequate. We can address their productivity.
It's not the problem for those that can't work--the truly disabled--but I think it is for those who can. They're not inherently incapable of being productive.
Impact on productivity is not the right way to judge a crime. Definitely not the sole measure or the paramount one. It would be open season on old people if that were the case. I suspect you wouldn't want to see that happen.
I don't believe the current system punishes offenders effectively or rehabilitates them. It's not great at segregating dangerous people either, as people are released when they've "paid their debt to society" rather than when they stop presenting a threat.
I'd prefer it if prisons were more like the prison farms in Scandinavia, especially for young non-violent criminals. They need to learn to produce things or render services that other prisoners will want. They should be able to build their own houses. They should have to learn how to work and freely trade with others BEFORE they are released.
Some criminals are a persistent threat to the population and must be tightly controlled. People who've murdered multiple people, for example. And some should never be released.
>understandings of genes as immutable features of biology that are fixed at conception
This 'understanding' has been known to be wrong for 30+ years now. Rearrangements, microchimerism, epigenetics, chromosome organization etc. are all things that exist and modulate information transfer beyond the standard four letters or the 'environment' (whatever people think it means) and we're still only scratching the surface. I keep saying it and I'll say it again: DNA is not source code, your genes are not initial character stats in a video game. That's not how any of this works and it frustrates me whenever I read otherwise smart people making arguments that rely on premises originating from a high schooler's (or 1970's) understanding of genetics. It's even worse when the dreaded 'nature/nuture' gets mentioned.
True. Most of what I've been taught after high school feels like "you've been lied to, everything you were taught in high school was wrong or grossly oversimplified and now you need to unlearn it all and relearn from scratch." Makes you wonder what the purpose of high school was in the first place.
I got the same impression from people who studied chemistry after high school too though. Some things are just simplified so our children can ramp up their understanding gradually instead of having the wild complexity of the universe thrown at them in one go.
It makes no sense to leave things out of a description of a complex system to help someone to understand it better.
Have you considered educators might not want 17 year olds to understand things that are "too dangerous"? Because that happens at a pretty basic level in chemistry.
> Makes you wonder what the purpose of high school was in the first place.
Oh, I have an opinion about that:
Well. First and above all the first purpose of school and high school is to learn to follow authority (edit: authority in every other meanings. Tthe first authority to submit oneself to is the school). This is the first brick and if it's not assimilated in a way that suits the school the student will get thrown out of the system.
A century ago there's the industrial revolution mantra: produce workers who can read and understand semi-complex or complex instructions in written and oral forms and manipulate tools (typewriters, screwdrivers, hydraulics, mechanics, machines, etc.).
And now there's the "learn to learn" mantra because few things taught in the 'regular', 'normal' or 'general' school and high school cursus land jobs in the post-industrial and the information societies. Does not seem to apply to trade schools.
This would also explain why we don't teach fundamentals of epistemology or critical thinking, even though we constantly read in the news how crucially important these skills are, because of fake news and Russian election hackers.
Last time I heard about my country in OECD ranking (or was it the PISA ranking ?) it was mentioned that 1 out of 4 students out of high school was almost illiterate. They can (barely) read but they don't understand the meaning behind the words.
Either it's a consequence of not teaching critical thinking or a reason we don't teach it is up to debate but the results remain.
> The first purpose of school and high school is to learn to follow authority.
I think that depends which person at the school you ask. I am married to a low-income neighborhood high school English teacher and we talk curriculum often. Seems to me that it's about striking a balance: One side is what you said, the other side is teaching students to unite against that authority in productive ways.
Too far in one direction, and you wind up with a society where none of the competent people you know have enough faith in the system to run for office. Too far in the other direction and the dominant class sees the everyone else as a threat and puts walls up.
It's a difficult balance to strike, but I know a lot of teachers that work very hard at finding it.
I want to make it clear that my thought is that the first authority to submit oneself to or to follow is actually the school. Even if the school puts emphasis on critical thinking you won't have a place in it if you don't submit to its authority.
It's not called the central dogma of biology for nothing!
But seriously, in my understanding it's taught that way to try to rid people of mystical pre-evolutionary notions. Once they have firmly grasped the basics they can move on to super-interesting corners where stranger things may happen, and about which people are still arguing.
The physics analogy is people arriving with mistaken intuition about "when does the dead body roll out of the truck" level problems. The fact that Newton isn't the end of the story doesn't spare you from having to un-learn Hollywood/Aristotle first. Nor will your earlier confusion be any help in wrapping your head around post-Newton concepts.
Every biology, molecular biology, and genetics textbook made for freshmen and sophomore bio majors. You can't pick a wrong one - this has been core material for at least two decades.
Really? So like, say, Campbell Biology? That's what I've seen used in high school and colleges, and I don't recall any mention of epigenetics in it. (At least in the 7th edition one, published 2004... not sure how much it's changed since.)
Confused, are we looking at the same edition? I don't see an explicit mention in the ToC [1], and Chapter 9 is "Cellular Respiration: Harvesting Chemical Energy"... are you saying that's where they mention epigenetics?
Yeah you're looking at the wrong book. That's copyrighted 2019, not 2004. And not sure what "7e" means (doesn't seem to be the same as "7th edition") but it's Campbell Essential Biology, not Campbell Biology.
So, the edition you’re linking, you’ll likely find epigenetics in ch21, genetic basis of development. That’s usually where the older books discussed methylation and acetylation.
They didn’t generally refer to it explicitly as epigenetics - they tended to cover the exact mechanisms. Check out pg 358’s dedicated subsection to methylation, for example. The word epigenetics won’t be mentioned, but it’s one of the classic epigenetic mechanisms.
I'm not sure if I understand that section correctly, but it sounds like it's talking not about inheritance across organisms, but about replication within an organism?
Methylation and acetylation of sex chromosomes is inherited. In somatic tissues it affects downstream development; in gonads, it becomes a heritable trait.
Genes do explain much of the variation between people for many characteristics. You can think of them as initial character stats that are then modified by environmental factors and random noise.
Again, that's not how our modern understanding of genomics works, and hasn't been since the 1980's. There's much, much more to it than genes+environment. See all the mechanisms I've mentioned in GP. Also, you brought up 'variation between people', which can be aditionnally confounded by things that have nothing to do with DNA at all, such as the various microbiomes.
It's important to note that heritability is not some kind of magic bullet that explains everything with "genes". If you want to convince a biologist that "genes" are the cause for a given phenotype, you've got to show an actual genetic causal mechanism or you've got nothing. This is, again, due to the myriad of interfering mechanisms that make "genes" very, very removed from the actual phenotype. And again, I'm not talking about the environment.
It's worth noting that there are two uses of gene here, or at least of "genetic".
One of which is roughly DNA that codes for a protein, or controls such, and indeed the causal mechanisms from there to phenotype can be extremely complicated and hard to figure out.
In the context of "variation between people" there's another meaning, of just whatever it is that causes inheritance. We could say many things about this (e.g. from twin studies) even before we'd discovered DNA, and non-DNA routes (like if most babies get some microbes from their mothers) would still count.
The complexity of how biological mechanisms work does not change basic facts about genetics. There isn't any conflict between the two.
Your second point is not true for complex polygenic traits like height. Height is strongly influenced by genetics but we will likely never understand the exact mechanisms that do this.
Also the compiler is self-modifying. And the compiler flags as well. And the OS. And the chip architecture sometimes. Like, at what point do we decide that pushing the analogy further is just silly and just drop it altogether?
I don’t disagree with you but I want to point out that it’s interesting how better software, running on an OS, running on a chip architecture, has enabled us to write better OSs and better chip which in turn allows us to write better software that allows us to repeat the cycle.
Aren’t we somehow simulating how life works? Higher abstractions/constructs improve the way the lower constructs work :). Of course, just an oversimplification.
And yet something guides the human zygote in developing as a human. At what point do we stop focusing on refuting simplistic ideas and turn our attention to what is now known?
That 'something' you mention is called developmental biology and it's yet another level of abstraction removed from the basic cell machinery that operates around DNA. It also has nothing to do with compilers or anything computer related. At some point the analogy is just more harmful than anything else.
At the end of the day, if the gene is self-modifying, the meta-code that modifies the gene also lies somewhere within the gene. Or are you talking about externally modifying a gene by inducing certain mutations?
I am no expert, but I have read a couple of good books on the topic and my understanding is that while it is true that your DNA can modify itself to some extent in response to stimulus, the extent to which this happens is very limited in scope, and limited to ancient genetic pathways, which means these get activated only in times of severe biological stress, like through starvation or famine.
I still don't buy the "There is no nature vs. nurture" statement. Maybe there's more than nature vs. nurture, or more complicated than nature vs. nurture. But totally no nature vs. nurture? Overstatement.
Nature vs. nurture hasn't been a thing since the 2000's when second generation sequencing dramatically expanded the breadth of information we had about species' genomes and (along with other discoveries made at the time) made us realize there was indeed a lot more to it than this dichotomy. It's about as relevant in our modern understanding of genomics as the debate between GOTOs vs. structured programming is to our modern understanding of software engineering.
No. It's not about whether there was "a lot" more. It's about how much it can add on top of "nature vs. nurture" to explain common social phenomena. It is very possible the new knowledge can be 100 times more complicated, but it only adds 1% more explanation power to everyday phenomena.
Take physics as an example, quantum physics has "a lot" more than classical physics, but no physicists will discard classical physics as being no longer valid, or no longer useful for explaining everyday physical world, or you should always use quantum physics in engineering, etc.
Explaining common social phenomena with modern human genomics isn't a thing period. If you'd attempted to introduce some sociobiology talking points in a typical genomics conference you'd be laughed out of the room. (That's assuming the organizers would even consider your abstract for a talk in the first place.) No one really discusses 'nature vs nurture' anymore outside of a few circles that somehow still haven't gotten the memo. It's just not relevant anymore to modern fields.
My, admittedly weak, understanding is that the analogy is (much) more apt than not. Sure, the means by which that code is 'compiled' and 'executed' is much different than in the computers we've built, but, for one, DNA, genes, chromosomes, and all of the other various levels of organization of those units of information seem to be remarkably, amazingly stable, for many (most? almost all?) organisms, which seems to strongly imply that 'DNA is source code' is about as true as 'Earth is a sphere'. Sure, they're both only approximately true, but how approximately they're true (or not) is very important. Asimov explained this extremely well:
What epigenetics is not is anything like Lamarckian inheritance, at least not at all to the same degree.
> Rearrangements, microchimerism, epigenetics, chromosome organization etc. are all things that exist and modulate information transfer beyond the standard four letters or the 'environment' (whatever people think it means) and we're still only scratching the surface.
Sure, but that doesn't mean that the central dogma of biology is entirely wrong, just that it's not always and everywhere true exactly. The current best understanding of epigenetics (quoted from my second link above) is that, whatever it is, it is such that:
> very little of [genetic] inheritance is perturbed by epigenetic effects
I'm not sure what exactly you're looking for, but yes, Craig Venter's team at JCVI and later Synthetic Genomics has been building synthetic organisms starting with Mycoplasma laboratorium in 2006. Most recently, an E. coli analog was built using DNA synthesis (https://www.nature.com/articles/s41586-019-1192-5).
To be clear, @aeorgnoieang's response is basically correct. The central dogma holds, the epigenetic program's role is well-characterized to be relevant to development/differentiation only under normal circumstances, epigenetic imprinting effects are weak, etc. As @aeorgnoieang correctly points out, the other things you cited (microchimerism, rearrangements, chromosomal organization/spatial effects) don't actually conflict with the "DNA as source code" metaphor in any way.
This is not to say that the popular understanding of these things is perfect - there are lots of misconceptions - but the way to fix that is by finding better ways to describe what we know and put it in context. A big issue is that as we learn more, media events around new discoveries make it hard to put the relative effects (of things like chimerism/mosaicism, polygenic risk factors, horizontal gene transfer, etc.) into context if you're not already an expert.
This paper is a PR gold mine but a methodological disaster, which is par for the course in "EWAS" (epigenome-wide association studies). At one point I had to remind a coauthor that pointing out structural problems in fashionable studies such as these is a quick way to lose (more) faith in modern peer review; for some reason she thought that EWAS enthusiasts gave a flying crap whether the "genes" they were studying were subject to structural variation (e.g. amplification and deletion). Ha! Ha! Ha!
Anyways. Their functional enrichment analysis is uncorrected for the known bias of the platform (something that has been repeatedly addressed by multiple authors since 2012), and no attempt appears to have been made to correct for cryptic stratification (i.e. structural polymorphisms, which are rampant in human populations, and particularly among so-called metabolic genes), though in the study population that may not be a major issue.
Quantile normalization is only appropriate if one can reasonably assert that the overall distribution of measurements is roughly the same between individuals and groups; this assumption has been shown to be invalid in the absence of positive and negative controls for gene expression, whence its original propagation, and more so for DNA methylation under various conditions. The batch correction approach used here is notorious for squashing real signal, although paradoxically that may have moderated some of the other methods choices.
Moreover, the paper demonstrates that a particular sample of high-SES vs. low-SES individuals in Cebu in the Philippines demonstrates some (fairly tiny) differences in DNA methylation at a relatively small number of CpGs (about 2000 out of 485000 or so measured and 110000 or so tested), without particular note as to whether the sites are clustered, functional, or otherwise of interest. The functional impact of these changes are difficult to interpret, partly because of the bias in the functional analysis (something that has been established for nearly a decade; the authors clearly went shopping for methods in a "confirmatory" style).
We shan't even bother to discuss the effect of [mal]nutrition on metabolism and thereby upon DNA methylation and cell composition (both intertwined, although an attempt was made to correct for the interaction), which further muddies the waters w/r/t SES as opposed to individual-level effects. The analysis is done with a fixed-effects model assuming unstructured shrinkage, which of course is a bit odd considering that the measurements have a relatively easily determined correlation structure (their sample size is sufficient to estimate this) and thus variance decomposition could have been highly informative. This is doubly odd for a population "epigenetics" study, given that variance components were literally invented in population genetics.
In conclusion, while it's a lovely piece for a PR department, the actual relevance of either the measurements or the phenomena to actual humans and public policy is quite difficult to interpret. Perhaps that was the point...
Everyone who wrote and authorized this press release ought to be fired. The paper offers no evidence of relevance to epigenetic inheritance, and has nothing to do with nature vs. nurture.
All that's going on here is that people in the Philipenes have differences in environmental exposures that affect gene expression in their immune system, and this, unsurprisingly, differs by SES. No evidence that any of these marks are more than temporary marks of current gene expression patterns let alone anything as shocking as passing through the germ line to the next generation.
> We did not find support for the hypothesis that low childhood SES would be associated with DNAm in young adulthood, independent of current SES. Prior research has reported sensitive periods of SES influence (Borghol et al., 2012; Lam et al., 2012), but our study is likely under-powered to test for this effect.
65 comments
[ 715 ms ] story [ 1853 ms ] threadIt's good to share links to previous threads, but only if there's actually a discussion there.
It is common to see why this could be a big problem for poor people as they have to worry about health, food, etc for day to day living while we take them as granted and can focus on other things like writing code, going to meetings, and the like without worrying about where the next meal is going to come from!
I think focusing on productivity would go further than the dichotomy's of "welfare or not", where people are mostly skeptical of whether a subsidized poor person is contributing to society with taxpayer's money.
Relegating the role of governance to productivity allows even prisoner rehabilitation to change. It allows jailable offenses to be viewed under the lens of whether this is useful to the productivity of society, instead of simply punishing someone.
Because that won't gain consensus.
But demonstrating how some parts of our culture undermine productivity can.
I think we could reach similar outcomes, if the lack of well-being can be proven to show that it undermines productivity.
I don't see anyone saying this, but I think fixing the bottom 10% of society requires fixing the top 90%. I see no way that happens. Nobody accepts blame for being wrong, and most people's cognitive capabilities are filled by an evening watching cable TV.
The bottom with disruptions to their income are detrimental to society’s health. Public sector monetary solutions are untolerated and private sector solutions of employment are often inaccessible or inadequate. We can address their productivity.
I don't believe the current system punishes offenders effectively or rehabilitates them. It's not great at segregating dangerous people either, as people are released when they've "paid their debt to society" rather than when they stop presenting a threat.
I'd prefer it if prisons were more like the prison farms in Scandinavia, especially for young non-violent criminals. They need to learn to produce things or render services that other prisoners will want. They should be able to build their own houses. They should have to learn how to work and freely trade with others BEFORE they are released.
Some criminals are a persistent threat to the population and must be tightly controlled. People who've murdered multiple people, for example. And some should never be released.
https://www.theatlantic.com/international/archive/2013/09/wh...
Anyway, a complicated topic for sure.
This 'understanding' has been known to be wrong for 30+ years now. Rearrangements, microchimerism, epigenetics, chromosome organization etc. are all things that exist and modulate information transfer beyond the standard four letters or the 'environment' (whatever people think it means) and we're still only scratching the surface. I keep saying it and I'll say it again: DNA is not source code, your genes are not initial character stats in a video game. That's not how any of this works and it frustrates me whenever I read otherwise smart people making arguments that rely on premises originating from a high schooler's (or 1970's) understanding of genetics. It's even worse when the dreaded 'nature/nuture' gets mentioned.
Have you considered educators might not want 17 year olds to understand things that are "too dangerous"? Because that happens at a pretty basic level in chemistry.
Oh, I have an opinion about that:
Well. First and above all the first purpose of school and high school is to learn to follow authority (edit: authority in every other meanings. Tthe first authority to submit oneself to is the school). This is the first brick and if it's not assimilated in a way that suits the school the student will get thrown out of the system.
A century ago there's the industrial revolution mantra: produce workers who can read and understand semi-complex or complex instructions in written and oral forms and manipulate tools (typewriters, screwdrivers, hydraulics, mechanics, machines, etc.).
And now there's the "learn to learn" mantra because few things taught in the 'regular', 'normal' or 'general' school and high school cursus land jobs in the post-industrial and the information societies. Does not seem to apply to trade schools.
Either it's a consequence of not teaching critical thinking or a reason we don't teach it is up to debate but the results remain.
I think that depends which person at the school you ask. I am married to a low-income neighborhood high school English teacher and we talk curriculum often. Seems to me that it's about striking a balance: One side is what you said, the other side is teaching students to unite against that authority in productive ways.
Too far in one direction, and you wind up with a society where none of the competent people you know have enough faith in the system to run for office. Too far in the other direction and the dominant class sees the everyone else as a threat and puts walls up.
It's a difficult balance to strike, but I know a lot of teachers that work very hard at finding it.
I want to make it clear that my thought is that the first authority to submit oneself to or to follow is actually the school. Even if the school puts emphasis on critical thinking you won't have a place in it if you don't submit to its authority.
But seriously, in my understanding it's taught that way to try to rid people of mystical pre-evolutionary notions. Once they have firmly grasped the basics they can move on to super-interesting corners where stranger things may happen, and about which people are still arguing.
The physics analogy is people arriving with mistaken intuition about "when does the dead body roll out of the truck" level problems. The fact that Newton isn't the end of the story doesn't spare you from having to un-learn Hollywood/Aristotle first. Nor will your earlier confusion be any help in wrapping your head around post-Newton concepts.
Chapter 9, pg 161 covers epigenetics, p 162 gene linkages, Ch11 covers gene methylation.
I’d find additional specific examples, but that’s all that’s laid out explicitly in the table of contents.
[1] https://www.pearson.ch/HigherEducation/BenjaminCummings/EAN/...
https://www.pearsonhighered.com/assets/preface/0/1/3/4/01348...
It's important to note that heritability is not some kind of magic bullet that explains everything with "genes". If you want to convince a biologist that "genes" are the cause for a given phenotype, you've got to show an actual genetic causal mechanism or you've got nothing. This is, again, due to the myriad of interfering mechanisms that make "genes" very, very removed from the actual phenotype. And again, I'm not talking about the environment.
One of which is roughly DNA that codes for a protein, or controls such, and indeed the causal mechanisms from there to phenotype can be extremely complicated and hard to figure out.
In the context of "variation between people" there's another meaning, of just whatever it is that causes inheritance. We could say many things about this (e.g. from twin studies) even before we'd discovered DNA, and non-DNA routes (like if most babies get some microbes from their mothers) would still count.
Your second point is not true for complex polygenic traits like height. Height is strongly influenced by genetics but we will likely never understand the exact mechanisms that do this.
It is source code, it's just that source code is self-modifying.
Aren’t we somehow simulating how life works? Higher abstractions/constructs improve the way the lower constructs work :). Of course, just an oversimplification.
I am no expert, but I have read a couple of good books on the topic and my understanding is that while it is true that your DNA can modify itself to some extent in response to stimulus, the extent to which this happens is very limited in scope, and limited to ancient genetic pathways, which means these get activated only in times of severe biological stress, like through starvation or famine.
Take physics as an example, quantum physics has "a lot" more than classical physics, but no physicists will discard classical physics as being no longer valid, or no longer useful for explaining everyday physical world, or you should always use quantum physics in engineering, etc.
My, admittedly weak, understanding is that the analogy is (much) more apt than not. Sure, the means by which that code is 'compiled' and 'executed' is much different than in the computers we've built, but, for one, DNA, genes, chromosomes, and all of the other various levels of organization of those units of information seem to be remarkably, amazingly stable, for many (most? almost all?) organisms, which seems to strongly imply that 'DNA is source code' is about as true as 'Earth is a sphere'. Sure, they're both only approximately true, but how approximately they're true (or not) is very important. Asimov explained this extremely well:
- [Asimov - The Relativity of Wrong](http://chem.tufts.edu/answersinscience/relativityofwrong.htm)
'Epigenetics' itself seems like it might not be a particular thing among different people:
- [The misunderstanding of epigenetics – Insitome](https://blog.insito.me/the-misunderstanding-of-epigenetics-a...)
What epigenetics is not is anything like Lamarckian inheritance, at least not at all to the same degree.
> Rearrangements, microchimerism, epigenetics, chromosome organization etc. are all things that exist and modulate information transfer beyond the standard four letters or the 'environment' (whatever people think it means) and we're still only scratching the surface.
Sure, but that doesn't mean that the central dogma of biology is entirely wrong, just that it's not always and everywhere true exactly. The current best understanding of epigenetics (quoted from my second link above) is that, whatever it is, it is such that:
> very little of [genetic] inheritance is perturbed by epigenetic effects
To be clear, @aeorgnoieang's response is basically correct. The central dogma holds, the epigenetic program's role is well-characterized to be relevant to development/differentiation only under normal circumstances, epigenetic imprinting effects are weak, etc. As @aeorgnoieang correctly points out, the other things you cited (microchimerism, rearrangements, chromosomal organization/spatial effects) don't actually conflict with the "DNA as source code" metaphor in any way.
This is not to say that the popular understanding of these things is perfect - there are lots of misconceptions - but the way to fix that is by finding better ways to describe what we know and put it in context. A big issue is that as we learn more, media events around new discoveries make it hard to put the relative effects (of things like chimerism/mosaicism, polygenic risk factors, horizontal gene transfer, etc.) into context if you're not already an expert.
Anyways. Their functional enrichment analysis is uncorrected for the known bias of the platform (something that has been repeatedly addressed by multiple authors since 2012), and no attempt appears to have been made to correct for cryptic stratification (i.e. structural polymorphisms, which are rampant in human populations, and particularly among so-called metabolic genes), though in the study population that may not be a major issue.
Quantile normalization is only appropriate if one can reasonably assert that the overall distribution of measurements is roughly the same between individuals and groups; this assumption has been shown to be invalid in the absence of positive and negative controls for gene expression, whence its original propagation, and more so for DNA methylation under various conditions. The batch correction approach used here is notorious for squashing real signal, although paradoxically that may have moderated some of the other methods choices.
Moreover, the paper demonstrates that a particular sample of high-SES vs. low-SES individuals in Cebu in the Philippines demonstrates some (fairly tiny) differences in DNA methylation at a relatively small number of CpGs (about 2000 out of 485000 or so measured and 110000 or so tested), without particular note as to whether the sites are clustered, functional, or otherwise of interest. The functional impact of these changes are difficult to interpret, partly because of the bias in the functional analysis (something that has been established for nearly a decade; the authors clearly went shopping for methods in a "confirmatory" style).
We shan't even bother to discuss the effect of [mal]nutrition on metabolism and thereby upon DNA methylation and cell composition (both intertwined, although an attempt was made to correct for the interaction), which further muddies the waters w/r/t SES as opposed to individual-level effects. The analysis is done with a fixed-effects model assuming unstructured shrinkage, which of course is a bit odd considering that the measurements have a relatively easily determined correlation structure (their sample size is sufficient to estimate this) and thus variance decomposition could have been highly informative. This is doubly odd for a population "epigenetics" study, given that variance components were literally invented in population genetics.
In conclusion, while it's a lovely piece for a PR department, the actual relevance of either the measurements or the phenomena to actual humans and public policy is quite difficult to interpret. Perhaps that was the point...
All that's going on here is that people in the Philipenes have differences in environmental exposures that affect gene expression in their immune system, and this, unsurprisingly, differs by SES. No evidence that any of these marks are more than temporary marks of current gene expression patterns let alone anything as shocking as passing through the germ line to the next generation.