"White Trash": the ugly reality and its import

I've been reading the recent book White Trash:  The 400-Year Untold History of Class in America, by Nancy Isenberg, Penguin Books edition, 2017.  This book documents the history of class divisions in our country, noting that it is not just racial minorities--native Americans and African Americans--who have been an integral part of our nation's history.  The white lower class, who have eked out their livings slaving away (one might say) for the wealthy elite, have been a part of the country since it's founding by European settlers.

The history of our 'democracy', including even our civil war, is one of gross oppression of the many by the few.  In post-WWII times, with industrial-scale materialism, we have distributed material goods and a better standard of living to a wide swath of the population, but the large underclass, that is unable to share in this, is still very much with us.  Somehow, it seems impossible to dispossess those who can afford yachts and private planes and multiple mansions, even enough just to even out the quality of lives of the many who still suffer deprivation here today.

This blog isn't social commentary as a rule, but one can ask about the degree to which genomic promises can be delivered nationwide, when living circumstances are so varied.  Among other things,  Big Data is in vogue, but how can there be large enough sample sizes to relate genetic variants to trait risk when, given so much disparity in access to resources, environmental variation is so great and so essentially unmeasurable?  If there are so many G and E combinations, then the assessment of their statistical correlation with various outcomes will be very shaky, at best.

Of course, in the nature of our history, white trash--the poor white population--was denigrated not just as being disgusting but also as genetically degenerate.  This idea fed the fervid eugenics movement, as Isenberg relates in some detail, which managed to rationalize the poor's situation as due to bad genes, casting the fear that they, being ethically unrestrained, will out-reproduce the rest of us--that is, the good guys.  This was of course just one of the convenient rationalizes that the elites always manufacture to justify the inherent worthiness of their privilege.  It has been thus since Plato, at least.  And we see much the same today, though usually implicit rather than crassly explicit.

The evil of social inequity, visited upon the many by the few, seems inevitably with us in history, and has led to a string of protest movements and even revolutions.  Whether equity is ever actually possible is something social scientists can ponder, but from a genetic point of view, there are some relevant issues, given what we've got.

Returning to bite: Sample size alone militates against biomedical/genomic equity
The promise of genomic miracles for "All of Us", the current slogan from NIH that promises to lead to precision medicine for everyone, sounds so laudable that its risibility may seem surprising. But such a promise is vacuous at best, and cynically self-serving of NIH. But the promise reflects some loss of memory by the Director of NIH.

Once upon a time the NIH, wanting some Big Money for some Big Data for genomics, proposed the HapMap project.  A smallish sample of a few tens of thousand individuals mainly from the majority populations (whites in the US, by some odd chance) would be genotyped for tens of thousands of markers across the genome, and these could be used for statistical association studies for, well, just about any trait you would want to name.

But wait a sec!  What about the needs of the many minority groups in the US and around the world?  How would or how could they benefit from this largely majority population sample?  The HapMap rational was that common diseases--the ones we really are targeting with association studies--were caused by common variants, and would hence apply to all populations; this was known as the CDCV 'hypothesis', though it was never justified on scientific grounds and was mainly a ploy to pry big long-term money out of NIH for the gene worshippers, and as some said quietly at the time, finally to pry funds from the grasping hands of environmental epidemiologists.

The problem, we now know, is that common diseases or even normal traits are due to countless individually minor, usually uncommon (low allele frequency) variants.  What we know now is that even for the 'major' populations (the whites and Asians with money for Big Pharma to collect), causation is not simple, not due to a couple of 'druggable' target genes, even if one should think of these common traits (diabetes, obesity, schizophrenia, IQ, etc) as typically genetic in the first place.

But let us here not be cynical about the implicitly racist or elitist aspects of NIH's Big Data schemes.  Let us think of how the failure of CDCV shows the hollowness of this year's NIH slogan, "All of Us", as any sort of real promise.

If (sample) size matters, than size matters!
Well, does size matter?  If not, if what we need are genetic variants common enough to be useful, then we can scrap the Big Data romance and go back to focused sample sizes in genetic studies.  This was the CDCV rationale.  But if that were true, why then today's push for massive samples, each person genome-sequenced?  Is this just a grab for money, or a way to hide behind a lack of actual ideas about how genomes work?

The nominal rationale is that to identify all the relevant causal genomic sites for a trait, we need very large sample sizes, since individual effects are by now known to be generally weak and rare in the population.  To find them, if statistical testing is our criterion, we need big samples.  That's convenient because it also means Big Data and regally funded projects too big and costly to terminate once started.

The CDCV rationale meant that modest sample sizes and sequence markers rather than whole sequence, would suffice to identify the genetic bases of the common later-onset diseases that we want to deal with (and, as a sort of freebie, other traits normal and otherwise, early and late onset).  And, conveniently, since these were common they would be shared among populations.  But that was wishful thinking.  If we are to include 'All' of us equitably, what would that mean?

First, how do you even devise a sample of "All' of us?  Given admixture and what that means for sampling, the inexact correlation between self-defined 'race' and gene history, differential accessibility of different populations, and so on, how do you get comparable representation of every subgroup in the US?  Whites are easy.  But should the same fraction of Cheyenne and Navaho and Mohawk and Cree be sampled, or just a fraction of "Native Americans"?  And this does not yet consider environments, socioeconomic status, and the like.  And, smaller populations mean fewer very rare variants, but at what point does the unavailability of sufficient samples mean less ability to detect relevant variables?

By the time CDCV failed to materialize, the justification for GWAS that it would be equitable became clearly false, but it was too late to stop the gravy trains of Big Data and 'omics-everything.

White Trash shows the persistent socioeconomic problem of inequity in our country.  It is about the sociology, and the sometimes biological rationales used to ignore or suppress the large fraction of the nation who are not among the yacht-owning privileged.  It discusses the eugenic and other pseudo-biological excuses that have been dreamed up by the privileged classes in the country, to justify or explain as inevitable and inherent in the poor.

GWAS proclamations, like "All of Us", are another reflection--perhaps unintentional, but nonetheless--of privilege and unfairness.  If traits really were genetic in the way being so widely promised (they aren't!), and even if we could accurately measure relevant environmental contributors (we can't), inequity is built into the statistical-significance nature of the game now being played because of sample-size issues that mean we cannot achieve comparable resolution among our different sub-populations.

Of course, clear secular trends in disease and other traits, like IQ and just plain height, show that genomes are not the major determining factor in the first place.  But who would want to give the money back to the greedy environmental epidemiologists?  And who would want to invest in real societal economic equity, to remove or reduce the disease-associated lifestyle factors that we know predominate?

Until we have a better way, we are in a sense in an ethical trap because genomics in its current form cannot deliver equity.  Environmental fairness would be more attainable, in principle, but human history has few if any real or lasting examples of that sort of societal equity.  The book White Trash makes the problem starkly clear, but provides no answers and perhaps rests on the idea that awareness of the problem might awaken some drive towards remediation.  Don't hold your breath.