N=1 drug trials: yet another form of legerdemain?

The April 30 issue of Nature has a strange commentary ("Personalized medicine: time for one-person trials," by Nicholas Schork) arguing for a new approach to clinical trials, in which individuals are the focus of entire studies, the idea being that personalized medicine is going to be based on what works for individuals rather than on what works for the average person.  This, we would argue, shows the mental tangles and gyrations being undertaken to salvage something that is, for appropriate reasons, falling far short of expectation, and threatening big business as usual.  The author is a properly highly regarded statistical geneticist, and the underlying points are clearly made.

A major issue is that the statistical evidence shows that many important and costly drugs are now known to be effective in only a small fraction of those patients who take them.  That is shown in this figure from Schork's commentary.  For each of 10 important drugs, the blue icons are persons with positive results, the red icons are the relative number of people who do not respond successfully to the drug.

Schork calls this 'imprecision medicine', and asks how we might improve our precision.  The argument is that large-scale sampling is too vague or generic to provide focused results.  So he advocates samples of size N=1!  This seems rather weird, since you can hardly find associations that are interpretable from a single observation; did a drug actually work, or would the person's health have improved despite the drug, e.g.? But the idea is at least somewhat more sensible: it is to measure every possible little thing on one's chosen guinea pig and observe the outcome of treatment.

"N-of-1" sounds great and, like Big Data, is sure to be exploited by countless investigators to glamorize their research, make their grant applications sound deeply insightful and innovative, and draw attention to their profound scientific insights.  There are profound issues here, even if it's too much yet another PR-spinning way to promote one's research.  As Schork points out, major epidemiological research, like drug trials, uses huge samples with only very incomplete data on each subject.  His plea is for far more individually intense measurements on the subjects.  This will lead to more data on those who did or didn't respond.  But wait.....what does it mean to say 'those'?

In fact, it means that we have to pool these sorts of data to get what will amount to population samples.  Schork writes that "if done properly, claims about a person's response to an intervention could be just as well supported by a statistical analysis" as standard population-based studies. However, it boils down to replication-based methods in the end, and that means basically standard statistical assumptions.  You can check the cited reference yourself if you don't agree with our assessment.

That is, even while advocating N-of-1 approaches, the conclusion is that patterns will arise when a collection of such person-trials are looked at jointly.  In a sense, this really boils down to collecting more intense information on individuals rather than just collecting rather generic aggregates. It makes sense in that way, but it really does not get around the problem of population sampling and the statistical gerrymandering typically needed to find signals that are strong or reliable enough to be important and generalizable.

While better and more focused data may be an entirely laudable goal, if quality control and so on can in some way be ensured, but beyond this, N-of-1 seems more like a shell game or an illusion in important ways.  It's a sloganized way to get around the real truth, of causal complexity, that the scientific community (including us, of course) simply have not found adequate ways of understanding--or, if we have, then we've been dishonorably ignoring what we know in making false promises to the public who support our work and who seem to believe what scientists say.

It's a nice idea, or perhaps one should say  'nice try'?  But it really strikes one as more wishful than novel thinking, ways to keep on motoring along with the same sorts of approaches to look for associations without good theoretical or prior functional knowledge.  And, of course, it's another way to get in on the million genome, 'precision medicine'© gravy train. It's a different sort of plea for the usual view that intensified reductionism, enumeration of every scrap of data one can find, will lead to an emerging truth.  Sometimes, for sure, but how often is that likely?

We often don't have such knowledge, but whether there is or isn't a conceptually better way, rather than a kind of 'trick' to work around the problem, is the relevant question.  There will always be successes, both lucky and because of appropriately focused data.  The plea for more detailed knowledge, and treatment adjustments, for individual patients goes back to Hippocrates and should not be promoted as a new idea.  Medicine is still largely an art and still involves intuition (ask any thoughtful physician if you doubt that).

However, retrospective claims usually stress the successes, even if they are one-off rather than general, at the neglect of the lack of overall effectiveness of the approach--as an excuse to avoid facing fully up to the problem of causal complexity.  What we need is not more slogans, but better ideas, questions, more realistic expectations, or really new thinking.  The best way of generating the latter is to stop kidding ourselves by encouraging investigators, especially young investigators, to dive into the very crowded reductionist pool.

Hyman Minsky, Charles Darwin, and descent into the cover of minutiae

The financial crisis was basically not predicted by our leading lights in the academic and intellectual economics community.  They had their very technical theories about how markets work, and how people behave economically--the rational, coolly calculating Homo economicus.  They had their 19th century and even earlier theoretical heroes, who are always cited.  There were somewhat differing schools of thought, but in fact they were, so to speak, more like different classrooms in the same building. Even with these differences, but they were alike in one thing: they were basically all wrong!  The wildly unstable speculation that led to the disaster of the 2000's was a policy result of this universal body of trusted advisors, Those Who Knew.

Well, not entirely.  There was a curmudgeonly economist named Hyman Minsky (1919-1996).  We're not economists and have only learned about him second-hand, after the fact, when what he said before the fact was born out by the facts.  A source we recently listened to was the BBC Radio program called Analysis (listen to or download the March 24 program).

Minsky; Levy Economics Institute

While fancy economists were building their mathematical 'models' of economic behavior, which were very intricate and detailed, ordinary people and the bankers who misled them were venturing hither and thither for the quick kills.  Minsky, basically out of the mainstream, was warning in less technical but actually far more relevant and correct ways that stability builds instability. As the Levy Economics Institute described his ideas in brief,

Minsky held that, over a prolonged period of prosperity, investors take on more and more risk, until lending exceeds what borrowers can pay off from their incoming revenues. When overindebted investors are forced to sell even their less-speculative positions to make good on their loans, markets spiral lower and create a severe demand for cash—an event that has come to be known as a "Minsky moment."

In the recent crisis, confidence in quick-profit investments was so great that people became careless and built their hopes and McMansions of sand. When what amounted to a grand, expanding Ponzi scheme finally collapsed, disaster struck for many (except those who could use the legal system to basically buy their way out of going to jail).

Minsky was just independent-thinking enough to be definitely out of what policy and university circles generally tolerate, and had died before the 2008 crash so he never saw his ideas vindicated.  They were subsequently adopted with post hoc enthusiasm, of course, by the very same prophets whose wisdom had led us to what actually happened (that is they didn't lose their university, bank, or think-tank jobs). Minsky is now apparently appearing with some prominence in new editions of economics textbooks (the idea of publishing books is perhaps a sign of total professional shamelessness, but that's another story).

On the radio discussion, the point was made that the Professionals, those Who Know have become ever more enamored of computer modeling, mathematical theory, simulations, and all the paraphernalia of technical 'science'.  In our highly risk-averse, technophilic, bureaucratized world, this passes for wisdom rather than soft-headed mainly verbal arguments (like Minsky's).  If you want to be published, get tenure or reach the next step on the think-tank or Wall Street status ladder, you better be very technical, and do things very narrowly and with elegant mathematics.  That that doesn't work, and it's known that it doesn't work, doesn't seem to matter ("well, it will work this time!").

This is a characteristic of our culture in our scientific age.  Reduction to technicality is what our institutions, reporters, governments, funders, advisors, and the like admire.  And that viewpoint has its tentacles elsewhere, too.

The same in evolution and genetics.
Like 19th century economists, Charles Darwin gave biologists their version of the truth.  It was a very broad theory, based on the traits of organisms.  This was what counted, not the underlying biological mechanism of the traits.  The argument was conceptual, with an implied quantitative basis.  Darwin actually viewed it as a mathematical theory much as Newton's theory of universal gravitation, but the mathematical details were unimportant.

Many scientists want to formalize such theory to give it support and the elegance of mathematics, but in fact, Darwin's own idea about the underlying basis ('gemmules' and 'pangenesis') was basically wrong.  Evolutionary theory proceeded well without any such basis and, indeed, today most biologists don't know or understand the mathematical claimant for the theory (called population genetics).

What the last 50 years have done is to attempt to reduce evolution to molecular and mathematical precision.  In particular, as genomic technologies have themselves evolved as dramatically as anything that ever happened to life, there has been a love-affair, or infatuation, with technology as if it were answering the basic questions about life.  Genetics does, indeed, illuminate many fundamentals about some aspects of life, but as we and many others have written extensively, it does not provide the global or precise kind of prediction that physics-envy would suggest.  Still, despite many facts being ignored or dismissed, such as the often poor predictive power from genotype to trait, contrary to the unstated causal assumption of genes as the fundamental 'instructions' of life, an enormous superstructure based on molecular and computer technology is being built on countless studies of minute details. Again, what we are seeing is reduction to technicality.

Hiding behind minutiae
Both areas shared the same sort of retreat to the depths of minutiae to establish their apparent profundity of understanding, wisdom, and influence.   Over-arching larger-scale understanding, rather unrelated to much of the minutiae, gets no attention: it's not technical and hence not glamorous enough. It sounds deeply important and so both the professions themselves and those who report their activities to the general public, and those who provide the funds for these activities, are impressed, buffaloed, intimidated, or otherwise persuaded.  But the diving into technical minutiae is a kind of bathos, that often does not seem to be much constrained by, or basically just bypasses, what we know and may even be obvious (as in economics).

These are just two areas in which one can draw some parallels.  They are undoubtedly widespread across many areas of our society, in science, semi-science, the arts and so on.  It does seem to be true that every culture has its traits, or themes, or belief systems.  In ours, it's a belief in technology and in particular computing technology.   Technology changes our lives, mainly for the better. But that it can solve many technical problems does not mean it leads to greater understanding.  Mathematics, despite Galileo's claim that it's the language with which God wrote the universe, is fantastically useful and precise when you can write equations whose assumptions are sufficiently accurate for your needs.  It can lead to outcomes that can be tested specifically.

But if the number and sorts of assumptions and structures (e.g., equations) that are constructed yield exact outcomes, those outcomes really are nothing more than the rewording of the assumptions.  That is, the deductions are contained within the assumptions and structures one choose to begin with.  There is no guarantee that the deductions represent the real world, unless the assumptions do.  Indeed, inaccuracies in assumptions and choice of structures can easily lead to unconstrained inaccuracies in the deductions, relative to the actual world.  The appearance of elegance and insight can be illusory even in theory.  (We might note here that the current kerfuffle over attempts to reinstate scientific racism also exemplify this kind of selective invocation of technical details or methods, while ignoring of more general countervailing facts that are well-known or obvious.)

This formal testability of mathematical predictions is often equated to--or confused with--proof of the assumptions on which it is based.  But they are assumptions, and if they are inaccurate your results will be precisely inaccurate.  Even matching predictions under such circumstances can, but need not, imply underlying truth.  This assumed to be causal can be correlated with what's truly causal, for example.

Further, when mathematical models and theories are thought to be precisely true--that is, assumed to be so--results from actual studies will rarely match predictions perfectly.  There will be human measurement and other technical errors, for example.  So how do we deal with these?  We use statistical or other sorts of tests, to judge whether the results match the predictions.  As we've written about before, we must rely on subjectively chosen tests of adequacy, like statistical significance level.  Superficial aspects of truth may pass such tests in a convincing way, but that doesn't mean the deeper, broader truths are being understood.

Worse than assuming that deviation of results from predictions are just technical errors, is the natural tendency to design studies and interpret results, in obliviousness to or willing ignoring of countervailing knowledge or facts. We do this all the time in science, even though we shouldn't.  Economists pretended everyone was a rational, perceptive value-calculating machine, when it was manifestly obvious that we are not.  Evolutionary geneticists assume Nature is a perfect screening machine, when it manifestly is not.

Verbal arguments can be global and true, but are not so easy to turn into specific predictions, hence their lower status than high-level technology. But ultimately science rests on verbal--conceptual--understanding.  Clearly in both economics and genetics (and who knows how many other fields?), we are in love with technology and use it for many reasons, delving deeper than our actual understanding allows.  Often that will generate findings or surprising facts that stimulate broader thinking, but just as often even scientists, enmeshed in the daily routine (rut?) of our careers,  have a hard time telling the difference.

We're human and we need our self-respect, sense of importance, salaries and retirement benefits, ego-stroking, and just plain sense that we are doing something of value and importance to our fellow humans.  We are all vulnerable to overlooking or circumventing deeper truths by hiding in minutiae that masquerade as truth, in order to attain those needs.  It happens all the time.  Usually, it doesn't matter very much.  But when misplaced claims of insight are uttered too charismatically, intercalate into too many societal vested interests, or are taken too seriously, then society can be in for a very rough ride to pay for its credulousness.  None of this is new, but if we are creatures who learn from experience, why don't we, or can't we, learn from our long history?

We are products of our culture.  One law of Nature may be that we cannot over-ride that law.

Red wine, resveratrol, and the credibility factor

On Tuesday we blogged about the latest Hot News that resveratrol, the long-established purportedly disease-protection ingredient in red wine and other foods, isn't protective after all.  The study was a modest-sized one in the Chianti region of Italy where the wine is good and the sunshine warm and welcoming.

Though not large or in any sense exhaustive, this study made the news outside of Vatican City only, we might suggest, because it seemed to contradict the literature that has for many years praised resveratrol, and red wine, for its benefits for supposed physiological or biochemical reasons.  Naturally, the news media were hot to seize on the new study.  It stirred controversy, a favorite of the media.

But how do we know which to believe?  Should this study cast doubt on the prior studies' results?  If it is a good and reliable result, one can see why it could be reported as News.  But hold on.

On what possible grounds would this study be viewed as definitive, or at least definitive enough, to cast doubts on the extensive prior work?  If a responsible reporter were contacted by the author or a journal about this, why should the reporter believe or even report these results?

Given the numerous studies supporting a protective effect for resveratrol, on what grounds would this be viewed as credible, or should the media assume there may be problems with the study and that possibility be the headline and basis of the the story?  One can think of many issues to do with the epistemology of this kind of study, such as confounding, including confounding socioeconomic factors, as we mentioned in our original post on this story.  Why isn't the first reaction of the media that there must be something wrong with the latest results, and the study's bottom line be incorrect?

Indeed,  why shouldn't the media treat this as a junk study, not worth reporting, since it did not seem to have any clear or definitive reason for countering prior results?  Or at least ask what other factors may have been correlated with exposure to resveratrol, to negate its normally reported effect?

One possibilities is that the result is just a statistical fluke, and the actual resveratrol effect in the Chianti population is protective but that the roll of the sampling dice by chance turned up an opposite result.  How would we know?

We can put this another way: doesn't the study show that the media, and perhaps some scientists, treat this as a more definitive study than the prior work, and that that is what they seem to assume makes it newsworthy?  One can ask whether we can't expect this one study to be used as a rationale for proposals for funding for yet more, larger studies to resolve this 'controversy'.  Or, should we not expect that next week or next year another study will appear, and be given credibility by the media and authors, that will reverse the current wisdom?

The idea that one factor has an important net effect that will be realized is the kind of reductionistic thinking we often criticize, including in our prior resveratrol post.  No author would admit to thinking in a one-factor way, because we all know that would make us vulnerable to criticism, but this story shows that underlying the work is just that sort of approach.  At some point, if this is the kind of result we get, and we don't assume the current study is flawed, we should say we know enough to conclude that resveratrol has a protective effect, and go forth with other studies rather than repetitions of this one.

Or, if for some reason we actually give credence to this story, we need to ask why so many prior studies were junk studies, and then the question arises as to who to believe, and when we actually have an answer--rather than just another rationale for not going on to something more important, or at least to questions that have actual answers.

Or, it's possible that both results are correct; resveratrol is protective in some populations and some contexts, and not in others.  As always, it's hard to know how to interpret contradictory results, and too often it's the latest results that are given most credence for no reason other than that they are the latest results.

Meanwhile, drink red wine if you like red wine,  but then don't drive, because that is a risk factor we should actually believe!

Not procrustean science

If blog posts were physical we'd have spilled a lot of ink trying to make the point that getting beyond uncertainty in most things in human affairs should be harder than it seems most people find it to be.  Science, social science, religion, philosophy, politics -- a close look at the foundations of what we think we know in any of these fields should at least be sobering.

Reductionism vs holism
A paper written by Carl Woese ten years ago, in 2004, "A New Biology for a New Century," is worth a look ten years later. Woese, who died about a year ago at the age of 84, was a leading microbiologist, the first to describe the archaeal kingdom of life -- the not-prokaryotes and not-eukaryotes.  He also argued strongly in favor of abundant horizontal gene transfer way back at the beginning of life.  That is, the not-common direct ancestor.  Viruses transfer 'horizontally' among contemporaries, rather than (or, in some cases as well as) from parent to offspring.  Chloroplasts and mitochondria are other examples that transferred into cells that joined, perhaps billions of years ago.  That is not like the classical Darwinian image of descent from common ancestors, though how common it was at the beginning of life is debatable, as well as how important to current life.

The bed of Procrustes; Wikimedia

Clearly, Woese wasn't averse to finding uncertainty where others didn't, challenging accepted wisdom.  And, that's what he did in his discussion of the state of biology in 2004 -- molecular biology as he saw it was too reductionist and too reliant on technology, and thus, in danger of focusing too closely not even on the trees, but on twigs on trees, or perhaps xylem and phloem.  His view was that it was becoming impossible to see the larger picture.  Or as he put it, "molecular biology could read notes in the score, but it couldn't hear the music."

According to Woese, 19th century biology faced two separate challenges -- understanding the gene and the cell, and understanding evolution and the nature of development and the organizational structure of organisms.  The first tasks could be, and were attacked in a reductionist way, by the corollary of the view in physics that reality only existed at the atomic level, and consisted of the interaction of atoms and whatever forces made that happen.  Using technology, then -- microscopes, PCR machines and DNA sequencers -- we learned a tremendous amount about genes and cells.  Cells were the reduced units of organs, and genes of cells, and understanding would be found at those reduced levels.

The larger, complex, holistic questions, on the other hand, are not amenable to technological or conceptual reductionism.  Indeed, Woese described complexity as "reductionism's nemesis" and he believed that biology continued to concentrate on reductionist approaches at great cost.

A heavy price was paid for molecular biology’s obsession with metaphysical reductionism. It stripped the organism from its environment; separated it from its history, from the evolutionary flow; and shredded it into parts to the extent that a sense of the whole—the whole cell, the whole multicellular organism, the biosphere—was effectively gone.

It's not that no one was addressing big, non-reducitive questions in the 20th century.  Anthropology, evolutionary psychology, economics, psychology, human genetics, and more, gave us the application of reductionist molecular biology to larger questions of a certain kind, so that we now know that how we vote or that we gamble are genetically determined, and that every trait has an adaptive purpose. And that can be traced in a usefully deterministic way to 'genes'.  This is often done with a kind of hubristic certainty that doesn't seem justified given how deep questions of complexity still are.

But reductionism has created a facile view of evolution, where adaptive Just-So stories are too easy to construct, hard if impossible to prove, and accepted as true, and for which single genes are implicitly given primal roles by the very nature of the reductionism involved.  What we've seen far too much of is simplistic answers to large questions that turn the questions into caricatures of themselves.  Evolution is reduced to simple linear stories like the one Holly critiqued the other day.  Traits are often poorly conceived or arbitrarily defined, and teleologically explained, a problem that Ken wrote about in a couple of posts (here and here) a few weeks ago.  Belief (uncritical acceptance of assumptions) becomes the foundation for too many choices about what's true and not true about the world.

In 2004 Woese wrote,

Darwin saw biology as a “tangled bank”, with all its aspects interconnected.  Our task now is to resynthesize biology; put the organism back into its environment; connect it again to its evolutionary past; and let us feel that complex flow that is organism, evolution, and environment united. The time has come for biology to enter the nonlinear world.

This hasn't been completely ignored, it must be said.  There are many thoughtful people tackling biological complexity.  But we know that complexity is stymying reductionist approaches to genetics and yet people keep trying to force the fit, and make reductionism the answer.

Procrustes
In classical Greek mythology, Procrustes was a criminal who produced an iron bed and made his victims fit the bed...by cutting off any parts of their bodies that didn't fit.  The metaphorical use of the word means"enforcing uniformity or conformity without regard to natural variation or individuality." It is in this spirit that Woese characterized much of modern biology as procrustean, because rather than adapt its explanations to the facts, the facts are forced to lie in a bed of theory that is taken for granted--and thus, the facts must fit!

Instead, a proper scientific theory makes minimal assumptions, things basically that can't be tested, and develops theory from them and the theory is tested against data.  In life sciences we too often assume what really should be tested and, formally or implicitly force the facts to fit.

The problem, of course, is that every fundamental that is challenged and tested--such as the nature of the evolution of traits of interest (e.g., the assumption that 'natural selection', and therefore 'genes' are and were responsible), without much more than tautological truth (that is, truth by assumption) provided as evidence.

It is a legitimate challenge to develop better criteria for adequately defining, assessing, evaluating, and explaining much of what we can observe in biology.

The empty organism: If not reductionism, what then?

There are some interesting recent parallels in what one may call the philosophy of science, that have to do with the issues related to determinism, reductionism, and our struggle to understand complex trait causation in terms resembling 'laws' of Nature.  We discussed a lead-in to this the other day.

If life is just a kind of fancy molecular biochemistry, and molecules obey fundamental, universal physical laws, then mustn't life also follow the same laws?  If not, what does that mean and what could be the evidence?  How could the laws be suspended?  And at what level would purely material, molecular/energy stop applying?

Answering these kinds of questions is problematic (because we have no actual answers), but nonethless shows an important way in which even asking the questions is not entirely about science but is also profoundly affected by sociocultural and historical circumstances.  These circumstances are complex, but have to do not just with the technology and methods that are available at any given time, but also with what is acceptable to think in the first place.

Gotcha! moments
Darwin and the excitement that his attempt at a universal physics-like theory of life (evolution by natural selection, implying rather strong genetic determinism) was an exciting event in science.  It threatened established scriptural religion, and its proponents felt highly empowered to rip religion based on faith in one kind of scripture for what in many ways amounted to faith in another accepted word--that of Darwin.

Since adaptive evolution of a natural kind (not that done in labs, or in agricultural breeding, or via pesticides and antibiotics, etc.) took place imperceptibly slowly and in the past, we must rely on indirect explanations and interpretations of the evidence.  Ever since Darwin, there have been widespread and rather hubristic declarations of selective stories about this trait or that--or, by some, the belief (and that's the right word for it) that basically everything in life is the result of specific adaptive selection.

More than that, along with striking research success in identifying genes, the historical belief developed that reductionism--ultimately, molecular explanations of everything.  Molecules are the sexy total truth of the world according to that view.  But such a view is not just objective science; instead, it also reflects society at large, as is very clear from the history of the life sciences since Darwin's time.

Gotcha! regains acceptability
History affects what is accepted or followed or believed.  Darwin stole explanations of life away from religion.  Among other things was the idea that what we are is inherited and is here only because it was adaptively successful in the past. This was a view of biological inherency.  As is well known, this immediately spawned the eugenics era.  The idea was that now that we (that is, elite scientists, mainly males) know the real truth of the nature of organisms, we can control rather than be controlled by Nature.  We can guide our evolution with this knowledge.  In a kind of extension or rebirth of the Utopianism of the 18th century, we could purge society of its ills and replace them with only that which is good.  Of course, now we were talking about people, not just social and governmental structures.  This means determining who reproduces and who does not, which is the ultimate value judgement that needed to be made if we were to imitate and speed-up the beneficent goals of Nature.  It might be harsh, if only some reproduce and others don't (or are prevented from it), but Nature itself is harsh, and so on.

This led beyond the rather piously benign idea that someone in authority would decide who could mate, to the less benign idea that someone in authority would decide who could survive.  Over several decades, this idea terminated (so to speak) in the Nazi death camps.  By the end of WWII, the eugenic view that who and what you were was dictated by your genes became so discredtied and distasteful that scientists developed a very different view.

This was the behaviorist or environmentalist view.  In psychology it was lead by BF Skinner in that period, especially in the US.  The idea was that what you are was based on your experience, not your inborn tools.  This was not a new idea, but genetic inherency had taken over as the prevailing view for nearly a century, and the view, whether informal or formal in regard to rejecting eugenics, was the environmentalist view.

In this view, reduction to genetics was not really thought to be of any use.  Whatever the mechanism or how brought about by genes, that (the brain, neurons, etc) internal stuff was just not relevant to understanding the traits--behavior, mainly--of the person.  Reductionism was not going to gain any insights, even if certainly the mechanisms must involve genes and nerves and so on.  It was even said that you could (or should) just assume that an organism was entirely empty inside!  We just need to look at the outside not the inside of our subjects.  We did not need to know anything about the insides to understand behavior, and trying to work out the way the complex wiring worked was a waste of time.  A recognition of the complexity of traits like behavior.  Indeed, for his time, Darwin had little alternative, but this--considering the trait, not the internal generative mechanism--was essentially what he studied in so much detail.

Cachet and cash, eh?
The evidence didn't change, but behavioral approaches and environmental determinism took over.  Indeed, the evidence isn't changing very much even now.  We know a bit about neuroscience that is relevant to behavior, but we're still not really explaining behavior in any serious sense by invoking this gene or that one.  But what is pursued, what people 'believe in' and get dogmatically excited about, and what is allowed or considered acceptable (whether for explicitly understood reasons or not) is changing.  As memory fades, and new practitioners replace the WWII generation, genetic determinism and inborn inherency are rapidly regaining respectability. There is simply too much cachet--and too much cash!--in 'modern' technical science, and too little revulsion at what we know has been done in the name of imposing value judgements by one group against another (using religion, science, or whatever else) as expedient justifications, for this reversal of what is acceptable to swing back the other way.

What we accept is not just based on hard-core science decisions, but to a huge extent depends on historical context as well.  Will it turn sour again?  The probability may be low but is certainly not zero, because elitist expert-based decisions on how society should be run (by them) for its betterment (as they see it) are just hard to keep down.

Of course, we are no more genetically determined, or not, than during Hitler and the prior eugenics times.  So there is no serious scientific reason for this swing back to earlier once-discredited ideas.  But it is now savory to believe in it, as higher-level analysis fails to answer questions (as it did before eugenics) we return to reductionistic inherency.  It just seems technological, real science, and it's lost its odor of abuse.  A new generation reinvents its beneficence for society.  The sexy tools (genome sequencers, fMRI scanners, and much more) are available and so are the grant funds and the journals hungering for The New Discovery (after all, should journalists remember the past any better than scientists do?).

If behavioral  and evolutionary psychology can't keep their hands off this potential societal dynamite, they're not alone by any means.  Genomics and other omics are beating on the same drum, assuming that traits like obesity must be understood not on their own terms but by looking 'inside' the organism to understand them.  Of course, there is and always has been reason for trying to understand how things work.  But there isn't enough understanding, not yet at least, for this to come nearly to what is being promised.

Still, at present, we haven't got good law-like alternatives.  Are there laws of how underlying mechanisms must work, how the determine complex traits, or whether in fact inherently probabilistic things may mean that reductionism simply cannot work very well for the kinds of explanations being sought.

At the very least, more circumspection is what is in order.

Where should reductionism meet reality?

The dawn of empiricism
The march of modern science began, roughly speaking, in the period about 400 years ago when modern observational science replaced more introspective theoretical attempts to characterize the world.  The idea was that we can't just imagine how the world must be, based on some ancient and respected thinkers like Aristotle (and the 'authority' of the Bible and church).  Instead, we must see what's actually out there in Nature, and try to understand it.

Designs for Leeuwenhoek's microscopes, 1756; Wikipedia

Another aspect of the empirical shift was the realization that Nature seemed to be law-like.  When we understood something, like gravity or planetary motion or geometry, there seemed clearly to be essentially rigid, universal laws that Nature followed, without any exception (other, perhaps, than miracles--whose existence in a way proved the rule by being miracles).

Laws?  Why?
Why do we call these natural patterns 'laws'?  That is a word that basically means rules of acceptable behavior that are specified by a given society.  In science, it means that for whatever reason, the same conditions will generate the same results ever and always and everywhere.  Why is the universe like that?  This is a subject for speculation and philosophers perhaps, because there is no way to prove that such regularities cannot have exceptions.  Nowadays, we just accept that at its rudiments, the material world is law-like.

What it is about existence that makes this the way things are is either trivial (how else could they be?) or so profound and wondrous that we can do no more than assert that this is how we find them to be.  However, if this is the case, and as evidence that new technologies like telescopes showed that classical thinkers like Aristotle had been wrong that the laws were so intuitive that we could just think about Nature to know them, then we need to find them outside rather than inside of our heads.  That way was empiricism.

The idea was that by observing the real world enough times and in enough ways, the ineluctable regular patterns that we could describe with 'laws' could be discovered. Empirical approaches led to experimental, or controlled, observation, but what should one 'control' and how are observations or experiments to be set up to be informative so we could feel that we knew enough to formulate the actual laws we sought?  As the history unfolded, the idea grew that the way to see laws of Nature clearly was to reduce things to the fundamental level where the laws took effect.  In turn, this led to the development of chemistry and our current molecular reductionism:  If absolutely everything in the material world (the only world that actually exists?) is based on matter and energy, and these are to be understood in their most basic, or particular (or basic wave-like) existence, then every phenomenon in the world must in some sense be predictable from the molecular level.

The alternative was, and remains, the notion that there are immaterial factors that cause things we observe in the material world.  Unless science can even define what that might mean, we must reject it.  We call it mysticism or fantasy.  Of course, there may be material things we don't know about, along with things we're just learning about (like 'dark' matter and energy, or parallel universes), but it is all too easy to invoke them, and almost impossible for that to be more useful than just saying 'God did it' -- useless for science.

If anything, reductionism that assumes that atoms and primal wavelike forces are all that there is could be like saying everything must be explained in terms of whole numbers, assuming that no other kinds of numbers (like, say, 1.005) exist.  But science tries, at least, to explain things as best we can in terms of what we actually know exists, and that, at present, is the best we can do.

But 'observe' at what level?
Ironically, this view of what science is and does doesn't help in some very similar ways.  That is the case for at least two primary reasons.

First, the classical view of things and forces is a deterministic one.  According to that, if we had perfect measurement, we could make perfect predictions.  Instead, it is possible or even likely that some things are inherently probabilistic.  Even with perfect observation, we can't make perfect prediction.  In what is actually not a true example but illustrates the point, even if we know which face of a coin is 'up' when we flip it, we can't predict how the coin will land.  All we can say is something like that half the time it will land with Heads up.

There is lots of debate about whether things that seem inherently probabilistic and hence each even not exactly predictable just reflects our ignorance (as it does in coin flipping!) or whether electron or photon behavior really is probabilistic.  At present, it doesn't matter: we can't tell so we must do our work as if that's the way things are.  One positive thing is that the history of science includes development of sampling and statistical theories that help us at least understand such phenomena.

But this means that reductionism runs into problems, because if individual events are not predictable, then things of interest that are the result of huge numbers of individually probabilistic events become inherently unpredictable except, at best, also in a probabilistic sense like calling coin-flips.  But with coins we know or can rather accurately estimate the probability of the only two possible outcomes (or three, if you want to include landing on the rim).  When there are countless contributing 'flips', so that the result is, for example, the result of billions of molecules' buzzing around randomly, we may not  know the range of possible outcomes, nor their individual probabilities.  In that case, we can really only make very general, often  uselessly vague, predictions.

Pallet of bricks; Wikipedia

Second, reductionism may not work because even if we could predict where each photon or electron might be, the organization of the trait we're interested in is an 'emergent' phenomenon that simply cannot be explained in terms of the components alone.  A building simply cannot be predicted from itemizing the bricks, steel beams, wires, and glass it is made of.

Complexity of the emergent sense is a problem science is not yet good at explaining -- and this applies to most aspects of life; e.g., we blogged about the genius of Bach's music as an emergent trait last week.    It, too, is something we can't understand by picking it apart, reducing it to single notes. In a sense, the demand or drive for reductionism is a struggle against any tendency to be mystic.  We say that yes, things are complicated, but in some way they must be explicable in reductionist terms unless there is a magic wand intervening.  The fundamental laws of Nature must apply!

Herein lies the rub.  Is this view true?  If so, then one must ask whether it is our current methods, that were designed basically for reductionist situations, need revision in some way, or whether some entirely new conceptual approach must rise to the challenge of accounting for emergent traits.

This seems to be an unrecognized but currently fundamental issue in the life sciences in several ways, as we'll discuss in a forthcoming post.

The sky is falling! (or not)

Now, here's a question prompted by an op/ed piece in the Sunday NYTimes: would chimps throwing darts at a dartboard predict a person's risk of disease as well as any GWAS results to date?  The op/ed piece is about the recent uproar in the political science world that arose when the US House of Representatives in May passed an amendment to a bill that would eliminate National Science Foundation funding for political science research.

The writer of the op/ed piece, Jacqueline Stevens, a professor of political science at Northwestern, says that her doomsaying colleagues will disapprove of her saying so, but that for once she agrees with this Republican initiative, even if it's motive is anti-intellectualism rather than any real understanding of the issue.  As she says, political science is spectacularly unable to predict major world events, and chimps throwing darts do just about as well, and millions and millions of dollars have been wasted on meaningless research.

...the government — disproportionately — supports research that is amenable to statistical analyses and models even though everyone knows the clean equations mask messy realities that contrived data sets and assumptions don’t, and can’t, capture.

That is, NSF rewards simplistic views of the world, and politically motivated views at that.  Or, said in terms we often use about genetics, political science has become a reductionist field, reducing complex events to single determining variables.  With no predictive power.  Stevens writes, "Many of today’s peer-reviewed studies offer trivial confirmations of the obvious and policy documents filled with egregious, dangerous errors."

And of course this doesn't apply to political science alone, even if Washington politicians are targeting just that field. It's true of economics, psychology, sociology, any aspect of social science that is reductionist and attempts to predict future events based on simple models. Economists and social scientists -- note the label -- aspire to being scientific.  If that meant careful factual analysis and proper data collection rather than views based on ideology, one might agree that there is something scientific about these fields.

But it has become a kind of dictum that one needs big mathematical or statistical models, computer data bases, and formal mathematical theories to be scientific.  Too often, if not typically, one either works with 'toy' models that are so stripped of detail as to be useless to the real world (despite the rationale that they tell us about the real world and where to look for effects), or they are so intricate that they snow people into believing that this is science because its arcane.

The humanities, religions, politics, and other areas of human endeavor have similar traits, but in this case we're talking largely about the kinds of social science that is done in universities, which are supposed to be about seeking truth rather than smoke-screening.  We doubt anyone can seriously argue that society is better off overall in terms of its psychological or social health than it was before the age of big research grants that are now the life blood of universities, and how social scientists justify their status and keep their jobs (teaching some, if and when they really have to).

Of course, while we think social sciences really do deserve all of this kind of critique, which is coming not from us but from their own ranks, MT readers will know that we certainly believe that genetics and medical sciences are somewhat comparable.  So, in fact, is physics with its mega-colliders, hype about life on alpha-centauri, and strings. Big science now is about building empires devoted to particular research strategies, exotic and impressively complex, despite knowing very well that they will not deliver what we promise. And the track record, overall, supports this. Research empires and establishments are made that work a certain technology or worldview, and they then are like oil takers, slow to change direction because they depend on continuity.  It's not really an evil so much as the normal way humans behave, especially when we do make at least some progress with societal benefit (perhaps much moreso, or more stably and accumulative in genetics than in social sciences), when we need to earn a living.

Research funding cuts or reform including some type of accountability for real results, not just CVs padded with long lists of publications might help.  Of course, the problems we want to solve in social and physical sciences alike are difficult so the failure to find easy answers isn't the issue. It's the failure to own up, the knowingly false promises, the need for grant continuity, and the amount of wasted public resources that's the problem.

Stevens offers this solution:

To shield research from disciplinary biases of the moment, the government should finance scholars through a lottery: anyone with a political science Ph.D. and a defensible budget could apply for grants at different financing levels. And of course government needs to finance graduate student studies and thorough demographic, political and economic data collection. I look forward to seeing what happens to my discipline and politics more generally once we stop mistaking probability studies and statistical significance for knowledge.

In any field, social or biological or physical, there are valuable kinds of data to collect. Census data, data on incomes, age and sex related aspects of well-being, and so on. And there are unsolved problems that ought to be soluble if we focus on them. What those more focused questions are in polysci, sociology, psychology, education, and economics is not for us to say. 

But in genetics, these involve traits that really are genetic in the usual, meaningful sense of the term. Huntington's disease, sickle cell anemia, muscular dystrophy and many others are examples: we know the responsible gene, even if other genes may contribute in minor ways. Genetic technologies should be able to do something about those disorders. It's where the funding should go.

Tomorrow, we'll comment on a related issue, the determination of behavioral scientists to prove that behavior is genetically determined and simple -- and that they will find the genes.

Reductionism, part II -- The Tunnel of Love

Choosing to wear blinkers
Yesterday, we suggested that even the early geneticists were well aware of the multifactorial causation of traits, and asked how the 'gene for' thinking that has driven so much recent research, largely without satisfying results, came to predominate as it does currently. Today we suggest that science restricts its view intentionally, as a pragmatic way of discovering the nature of aspects of Nature. Indeed, the early geneticists did the same. And we point out that the price we pay is the way scientific methods restrict the degree to which we understand things more broadly.

'Gene-for' thinking is pragmatic -- understanding the molecular basis of genes and how they work is easier than understanding, say, polygenic interaction or the effect of the environment on development.  And of course great progress was made in molecular genetics throughout the 20th century, which only reinforced the view that the molecule was the thing. This, coupled with formal population genetics, gave researchers rules (how genes segregate, how DNA codes for proteins, and so forth) for cataloging how genes work, giving the field a theoretical framework within which to plan, execute and interpret experiments.

Discoveries in other fields sometimes reinforced the determinist view as well. Following not long after the 'one gene one enzyme' dictum was hypothesized by Beadle and Tatum in 1941, e.g., the coming of the computer age underscored the view of genes as the program or blueprint for life, an appealing and seductive idea that is yet to die. Even if a blueprint needs an architect, and a foreman to supervise the building.

In fact, the idea that the early geneticists had a broader view is only partially true. That they did is well-documented, as we wrote yesterday, but it wasn't ever really put into practice. Then as now, experiments were conducted in a way that enabled these guys to find single genes that 'caused' the traits they were interested in; environment was controlled, and fruit fly lines homozygous for a trait known to be due to the effects of a single gene were crossed so that the effect of a given gene could be assessed, just as Mendel had done with his pea plants.

This is how Morgan mapped genes. And, why those genes were given names like 'hairy wing', 'small eye', 'small-wing', 'vermilion', as though they were the single cause of or were 'for' these traits, even though Morgan knew full well that wing characteristics or eye color were due to many genes. Indeed, he wrote in The Theory of the Gene, "... it may appear the one gene alone has produced this effect. In a strictly causal sense this is true, but the effect is produced only in conjunction with all the other genes."

The question thus becomes a philosophical one about causation. Philosopher of science Ken Waters has written a nice paper about this*, discussing the difference between 'potential' and 'actual difference makers' and how experimental method determines which is found, while prior assumptions determine which are sought. Although Morgan knew that it took many genes to change eye color in flies -- potential difference makers, in Waters' terminology -- the gene that actually changed eye color in his experiments, a direct consequence of the way he conducted them, was the 'vermilion' gene. The actual difference maker. The foundation for gene-for thinking was well-established right from the beginning, and reinforced all along the way.

And, of course, the idea that some of the early eugenicists may have understood that environmental influences could be important in development didn't prevent the Nazis from making life and death decisions based on heredity.

'Gene-for' fervor takes off -- and people actually believe it
After the discovery of the gene for cystic fibrosis in the late 1980s, genes for more than 6000 single-gene disorders were quickly identified. These are largely rare, pediatric diseases, but even so there seemed to be little reason to assume that geneticists wouldn't continue finding genes for disease, and then even for behavior and other kinds of 'normal' traits, even if an important aspect of pediatric disorders is that they occur near birth and hence are relatively less susceptible to environmental effects (not entirely, of course, because even the uterine environment can vary).

This kind of success at finding genes associated with traits was seductive, and the commitment to strong genetic determinism is now found not only among geneticists, but among epidemiologists, psychologists, economists, political scientists, and even further afield. Epidemiology, e.g., had its own history of success finding the cause of infectious diseases, as well as the effects of environment risk factors like asbestos or smoking. But, as with single-gene disorders, when the effect of a risk factor is large, it's a lot easier to find than when there are many cumulative risk factors, some genetic and some environmental. When epidemiology turned to common chronic conditions like heart disease, asthma or diabetes, which generally don't have a single strong cause, they ran into the same kinds of epistemological and methodological difficulties that geneticists were having with these same complex diseases.

Ironically, out of frustration with the difficulty of finding environmental causes for many chronic diseases, epidemiology turned to genetics, and the field of genetic epidemiology quickly grew. Only to be as stymied in terms of the fraction of cases explainable by known genes. The 'strictly numerical basis' upon which Morgan had identified so many putative genes was no longer good enough. Because the counts don't come out in Mendelian terms unless fudge factors (called 'incomplete penetrance', a determinist idea itself, as it imbues the gene with a mystical ability to be more or less expressed) are added to account for other causes relative to a gene under study, usually meaning the gene accounts for only a small fraction of cases and doesn't have nearly Mendelian ratios among siblings, etc.

And it becomes institutionalized
Then of course when the Human Genome Project was finished, the sequencing factories had to be kept running, so yet more promises were made about what we were going to be able to do with genes, more billions were spent on more classically reductionist 'count only' genetics -- and yet the same problems remain unsolved. We still can't enumerate the genes that are responsible for height, and for exactly the reasons Morgan spelled out in 1926, as we noted yesterday.

In fact, the scientific methods that we use identify genes that, when mutated in some ways, cause serious stature problems (Marfan's syndrome makes you very tall, many genes make you very short), but when we look at the normal range as seen in a sample of healthy people, these genes do not generate mapping 'hits' (as in GWAS association studies). And this is probably true of most traits -- it's easier to explain the extremes of their distribution than it is to explain the normal range.

It's not that genes are unimportant. It's that we are only taking into account part of the truth. This is driven essentially by methodological considerations -- we've got well-developed formal theory for genes and how they segregate in families, and what that means about how to find them. But it is more of a struggle to account in useful ways for complex causation -- useful, at least, in terms of dreams of miracle drugs or genetically focused personalized predictions.

Tunneling through the truth

An important reason for the combination of great success in discovery in genetics, and the relatively great failure to account for complex traits has to do with methodology rather than the state of Nature. As we said yesterday, the focus on fixed, chromosomally localized causal elements -- 'genes' in the classical sense -- was driven by Mendel's careful choice of experimental material, followed by similar constraints employed by Morgan and the other classical geneticists of the early 20th century.

The very same logic and approaches have been followed to this day. Genes are identified as localized causal elements in DNA and we study and manipulate them through variation that is studied, as much as possible, by removing all other sources of variation. Traits are narrowly defined, transgenic experiments use inbred animals manipulated one gene (or one nucleotide) at a time, and so on. This is done because it generates a cause-effect situation that is tractable.

That approach, or research program, led to the steady discovery of the nature of DNA, of genes as protein codes, and so on, up to the mapping of entire genomes of a rapidly growing number of species.

Yet at the same time, when it comes to complex traits, we know that we are not discovering the whole truth -- and we know why. It is the same control of variation that led to discovery, that leads to obscuring the whole nature of Nature.

In a sense, what science does is to 'tunnel' through reality. Like any other tunnel, the walls are reinforced to keep things outside the tunnel out, and to make a clear path within. The path is a particular gene we are interested in, and we manipulate that gene, and its variation, treating it as a cause, to see what effects it has. We know it really interacts with the world outside, but we standardize that world as much as we can, to reveal only the effect of variation in the single cause.

This is perhaps a tunnel of love of experimental design, but not so much of the nature of Nature, because by particularizing findings, even on a large scale, we systematically isolate components from each other whose true essence, and origins, are intimately dependent on their interactions.

Maybe tunneling through truth is the only way science can understand the world. From the point of view of garnering facts, and manipulating the world by manipulating the same facts, science is a huge success. But in terms of understanding Nature, maybe we need a different way. If so, as long as the reductionist legacy of the 300 year old Enlightenment period in human history lasts, we will remain the Tunneling species.

Tomorrow we'll discuss how the same kind of thinking has worked in developmental genetics and the EvoDevo world of research.

-------------------------
*The Journal of Philosophy is only available online to members, but the reference is Waters, "Causes That Make a Difference", The Journal of Philosophy, 104:551-579, (2007).

Personalized archaeology

A young archaeologist gave a talk in our department last week about his excavation of rock shelters in the north east United States. He tries to deduce from artifacts such things as how much time people spent in each shelter, whether they hunted nearby and what it was they got, whether they returned to the same site year after year, caching hunting implements such as flaked or bola stones there for future use, changes in use of the site over millennia, and so on.  (Image by Larry D. Moore, used under a Creative Commons ShareAlike License.)


Several things were noteworthy, at least to some of the non-archaeologists in the audience. First, he talked about how differences in the construction of projectile points on the east coast versus those found in other sites in the Americas contribute to the idea that North America may not have been peopled entirely by migrants that crossed the Bering Straits, but that some may have come from what is now France. (Called the 'Solutrean hypothesis', after a French archaeological site called Solutré this idea posits hunter-gatherers traveling along an Arctic ice shelf across the Atlantic, either on foot or by canoe. The archaeological and genetic support for this idea is weak to non-existant.)


And second, he contended that his practice of dividing his study sites into smaller sections than the usual (his are 30cm x 30cm x 5cm) yields information at the level of the individual. That is, he believes that by restricting the size of each plot he excavates, and knowing the average human reach radius and so on, he's able to deduce how individuals spent their time in the shelter, shaving the points they'd use to hunt the next day, or cooking the day's catch and so forth.


That could be interesting -- like reconstructing campfire gossip about who was where when, and what they were doing.


But, really, it's hard to figure how it could be much more than that, something that an archaeologist I spoke with after the talk said as well. As a science, archaeology is about synthesizing observations into generalizations, to test or develop theory about human behavior. Just as any science.  This requires many observations, spanning long time periods and many different sites. (Although, certainly, as in any science, some archaeologists are not so interested in generalizing, and campfire gossip is enough.)


Later that day, I happened to be reading an old critique of the Human Genome Project, and I stumbled across the following paragraph:

[The HGP] is a powerful strategem to answer only certain peculiar questions relevant to its narrow purview. In summary, our critique is based on the following assessment: (i) going to the lowest level of organization does not necessarily yield any insight of interest; (ii) reductionist explanation, even when possible, is not cost-effective in terms of effort expended; (iii) mapping is justified, blind sequencing is not; and (iv) the sheer complexity of a system might make reductionist explanation impossible. (Tauber and Sarkar, The ideology of the human genome project, J R Soc Med. 1993 September; 86(9): 537–540.)

This could just as easily be describing the reductionist approach of our young archaeologist -- or indeed reductionism in general. What does it tell you to know that someone sat exactly here in the rock shelter sharpening a point? Or even that he was eating roast rabbit as he did so. (This is assuming, of course, that all the methodological issues that could prevent drawing such conclusions, such as dogs making away with animal bones, or burrowing rodents disturbing the dating information contained in the layering of the artifacts in the soil, and so on, were taken care of.) It certainly can't answer broader questions such as how the Americas were peopled, or how long ago that was.


In its race to reduce normal traits as well as behavior, disease, or risk of disease down to the level of the gene, modern genetics has turned the usual scientific method on its head in some ways, rather like reconstructing the activities of an individual at the campfire. And the torrents of sequence data that have been pouring out of labs around the world have led to 'hypothesis-free' analysis. Researchers now comb the data looking for interesting patterns, or the 'gene for' a trait, rather than for support for an hypothesis.


In this kind of thinking, we are still prisoners of Mendel, reducing our explanations to single genes, rather than accepting what has been known for at least 100 years, that most traits are polygenic, and have a strong environmental contribution. Indeed, even the seminal text on genetics and 'racial hygiene', which helped fuel the eugenic era of Naziism and the Holocaust, first published in 1921 and followed by numerous revisions, Human Heredity by Baur, Fischer and Lenz, explicitly recognizes the role of the environment.


But the seduction of genetic determinism and reductionism remains strong and powerful, in spite of the evidence. 

Everything's connected to everything else

A study out of the University of Indiana, described on the BBC website today, reports that married people have better cancer survival than unmarried or those going through the stress of divorce, etc. This is apparently not related to the level or quality of treatment.

We note this report not because it is particularly singular on its own, but because it is one of countless studies, including placebo effects, that show the interconnectedness of our physiological systems. 'Psychosomatic' effects, often sneeringly denigrated, clearly occur. Genetically, there are connections between neural, immune, and endocrine systems. These may be the flags that show why such otherwise non-physical effects exist.

For research the implications can be profound. For many reasons, practical, cultural, and historical, reductionist approaches are at the heart of the current modus operandi of science. Maybe there are other methods, but for 300 years or so these have transformed science and technology and are deeply ingrained into our thinking and science training.

However, in this and many other ways, reductionist methods may draw attention too far down the causal chain. It could focus us on trying to isolate causes that can't be isolated relative to the function we're studying. A gene can be studied on its own perhaps, but if its interactions with other genes are what we're studying, our methods need to reflect that.

It's common and easy to criticize reductionism, and we're not doing that here, per se. However, old habits die hard. We've known ever since Darwin that all life goes back to a singular beginning, and genomic sciences show that rather clearly in terms of DNA mechanisms, sequence variation, and so on. What recent sequence data have shown is that mechanisms long ago established can be highly conserved over long evolutionary times. Traits long thought to be independently evolved are regularly found to be homologous in various ways, both deep and subtle.

Thus our attempt to parse systems into discrete modular subsets is ultimately doomed to be inexact. When and how a study of, say, immunogenetics, can be done on its own without considering, say, neurobiology, is not clear. We can't design single studies to cover everything, and reductionism is needed in any kind of study (you can't study genes without studying genes!). Similarly, holism is a term that often sounds clearer than it really is.

Evolution didn't care: what worked proliferated. Evolution by phenotype sorts organisms by their results, not caring about their specific causes. Since life today is due to common ancestry, the problem we see is in no way a surprise. But there probably should be more explicit and serious studies of how and where to 'reduce' and how to integrate.