Modular complexity, and reverse engineering the brain

The Forbes columnist Matthew Herper has a profile of Microsoft co-founder Paul Allen who has placed his bets on a brain institute whose goal is to to map the brain...or at least the visual cortex. His institute is engaged in charting the sum total of neurons and other working parts of the visual cortex and then mapping their connections. Allen is not alone in doing this; there's projects like the Connectome at MIT which are trying to do the same thing (and the project's leader Sebastian Seung has written an excellent book about it) .

Well, we have heard echoes of reverse engineered brains from more eccentric sources before, but fortunately Allen is one of those who does not believe that the singularity is near. He also seems to have entrusted his vision to sane minds. His institute's chief science officer is Christof Koch, former professor at Caltech and longtime collaborator of the late Francis Crick who started at the institute this year. Just last month Koch penned a perspective in Science which points out the staggering challenge of understanding the connections between all the components of the brain; the "neural interactome" if you will. The article is worth reading if you want to get an idea of how simple numerical arguments illuminate the sheer magnitude of mapping out the neurons, cells and proteins that make up the wonder that's the human brain.

Koch starts by pointing out that calculating the interactions between all the components in the brain is not the same as computing the interactions between all atoms of an ideal gas since the interactions are between different kinds of entities and are therefore not identical. Instead, he proposes, we have to use something called Bell's number B(n) which reminds me of the partitions that I learnt when I was sleepwalking through set theory in college. Briefly for n objects, B(n) refers to the number of combinations (doubles, triples, quadruples etc.) that can be formed. Thus, when n=3 B(n) is 5. Not surprisingly, Bn scales exponentially with n and Koch points out that B(10) is already 115,975. If we think of a typical presynaptic terminal with its 1000 proteins or so, B(n) already starts giving us heartburn. For something like the visual cortex where n= 2 million B(n) would be prohibitive. And as the graph demonstrates, for more than 10^5 components or so the amount of time spirals out of hand at warp speed. Koch then uses a simple calculation based on Moore's law in trying to estimate the time needed for "sequencing" these interactions. For n = 2 million the time would be of the order of 10 million years.

And this considers only the 2 million neurons in the visual cortex; it doesn't even consider the proteins and cells which might interact with the neurons on an individual basis. Looks like we can rapidly see the outlines of what Allen himself has called the "complexity brake". And this one seems poised to make an asteroid-sized impact.

So are we doomed in trying to understand the brain, consciousness and the whole works? Not necessarily, argues Koch. He gives the example of electronic circuits where individual components are grouped separately into modules. If you bunch a number of interacting entities together and form a separate module, then the complexity of the problem reduces since you now have to only calculate interactions between modules. The key question then is, is the brain modular? Commonsense would have us think it is, but it is far from clear how we can exactly define the modules. We would also need a sense of the minimal number of modules to calculate interactions between them. This work is going to need a long time (hopefully not as long as that for B(2 million) and I don't think we are going to have an exhaustive list of the minimal number of modules in the brain any time soon, especially since these are going to be composed of different kinds of components and not just one kind.

Any attempt to define these modules are going to run into problems of emergent complexity that I have occasionally written about. Two neurons plus one protein might be different from two neurons plus two proteins in unanticipated ways. Nevertheless this goal seems far more attainable in principle than calculating every individual interaction, and that's probably the reason Koch left Caltech to join the Allen Institute in spite of the pessimistic calculation above. If we can ever get a sense of the modular structure of the brain, we may have at least a fighting chance to map out the whole neural interactome. I am not holding my breath too hard, but my ears will be wide open.

Image source: Science magazine

Physics, biology and models: A view from 1993 and now

Twenty three years ago the computer scientist Danny Hillis wrote an essay titled "Why physicists like models, and why biologists should." The essay gently took biologists to task for not borrowing model-building tricks from the physicists' trade. Hillis's contention was that simplified models have been hugely successful in physics, from Newton to Einstein, and that biologists seem to have largely dismissed model building by invoking the knee-jerk reason that biological systems are far too complex.

There is a grain of truth in what Hillis says about biologists not adopting modeling and simulation to understand reality. While some of it probably is still a matter of training - most experimental biologists do not receive training in statistics and computer science as a formal part of their education - the real reasons probably have more to do with culture and philosophy. Historically too biology has always been a much more experimental science compared to physics; Carl Linnaeus was still classifying animals and plants while Isaac Newton was mathematizing all of classical mechanics. 

Hillis documents three reasons why biologists aren't quick to use models.

For various reasons, biologists who are willing to accept a living model as a source of insight are unwilling to apply the same criterion to a computational model. Some of the reasons for this are legitimate, others are not. For example, some fields of biology are so swamped with information that new sources of ideas are unwelcome. A Nobel Prize winning molecular biologist said to me recently, "There may be some good ideas there, but we don't really need any more good ideas right now." He might be right. 

A second reason for biologists' general lack of interest in computational models is that they are often expressed in mathematical terms. Because most mathematics is not very useful in biology, biologists have little reason to learn much beyond statistics and calculus. The result is that the time investment required for many biologists to understand what is going on in computational models is not worth the payoff. 

A third reason why biologists prefer living models is that all known life is related by common ancestry. Two living organisms may have many things in common that are beyond our ability to observe. Computational models are only similar by construction; life is similar by descent.

Many of these reasons still apply but have evolved for the better since 1993. The information glut has, if anything, increased in important fields of biology like genomics and neuroscience. Hillis did not live in the age of 'Big Data' but his observation precedes it. However data by itself should not preclude the infusion of modeling; if anything it should encourage it even more. Also, the idea that "most mathematics is not very useful in biology" pertains to the difficulty (or even impossibility) or writing down, say, a mathematical description of a cat. But you don't have to always go that far in order to use mathematics effectively. For instance ecologists have used simple differential equations to model the rise and fall of predator and prey populations, and systems biologists are now using similar equations to model the flux of nutrients, metabolites and chemical reactions in a cell. Mathematics is certainly more useful in biology now than what it was in 1993, and much of this resurgence has been enabled by the rise of high speed computing and better algorithms. 

The emergence of better computing also speaks to the difficulty in understanding computational models that Hillis talks about; to a large extent it has now mitigated this difficulty. When Hillis was writing, it took a supercomputer to perform the kind of calculation than you can now do on your laptop in a few hours. The advent of Moore's Law-enabled software and hardware has enormously enabled number-crunching in biology. The third objection - that living things are similar by descent rather than construction - is a trivial one in my opinion. If anything it makes the use of computational models even more important, since by comparing similarities across various species one can actually get insights into potential causal relationships between them. Another reason Hillis gives for biologists not embracing computation is because they are emotionally invested in living things rather than non-living material things. This is probably much less of a problem now, especially since computers are used commonly even by biologists to perform routine calculations like graphing.

While the problems responsible for biologists' lukewarm response to computational models are legitimate, Hillis then talks about why biologists should still borrow from the physicists' modeling toolkit. The basic reason is that by constructing a simple system with analogous behavior, models can capture the essential features of a more complex system: This is in fact the sine qua non of model building. The tricky part of course is in figuring out whether the simple features truly reproduce the behavior of the real world system. The funny thing about models however is that they simply need to be useful, so they need not correspond to any of our beliefs about real world systems. In fact, trying to incorporate too much reality into models can make them worse and less accurate.

A good example I know is this paper from Merck that correlated simple minimized energies of a set of HIV protease inhibitor drugs to their inhibition values (IC50s) against the enzyme. Now, nobody believes that the very simple force field underlying this calculation actually reproduces the complex interplay of protein, small molecules and solvent that takes place in the enzyme. But the point is that they don't care: as far as the model is predictive it's all good. Hillis though is making the point that models in physics have been more than just predictive, they have been explanatory. He extols biologists to move away from simple prediction, as useful as it might be, and towards explanation. 

I agree with this sentiment, especially since prediction alone can lead you down a beatific path in which you may get more and more divorced from reality. Something similar can happen especially in fields like machine learning, where combinations of abstract descriptors that defy real world interpretation are used merely because they are predictive. These kinds of models are very risky in the sense that you can't really figure out what went wrong if their break down; at that point you have a giant morass of descriptors and relationships to sort through, very few of which make any physical sense. This dilemma is true of biological models as a whole, so one needs to tread a fine line between keeping a model simple and descriptive and incorporating enough real world variables to make it scientifically sensible.

Later the article talks about models separating out the important variables from the trivial ones, and that's certainly true. He also talks about models being able to synthesize experimental variables into a seamless whole, and I think machine learning and multiparameter optimization in drug discovery for instance can achieve this kind of data fusion. There is a twist here though, since the kinds of emergent variables that are rampant in biology are not usually seen in physics. Thus, modeling in biology needs to account for the mechanisms underlying the generation of emergent phenomena. We are still not at the point where we can do this successfully, but we seem to be getting there. Finally, I also like to emphasize one very important utility of models: as negative filters. They can tell you what experiment not to do or what variable to ignore or what molecule not to make. That at the very least leads to a saving of time and resources.

The bottom line is that there is certainly more cause for biologists to embrace computational models than there was in 1993. And it should have nothing to do with physics envy.

Three reasons why nobody should be surprised by junk DNA

Science writer Carl Zimmer just wrote an article in the New York Times about the junk DNA wars and it's worth a read. Zimmer presents conflicting expert opinions about the definition of 'junk' and the exact role that junk DNA plays in various genetic processes. As he makes clear, the conversation is far from over. But personally I have always thought that the whole debate has been a bit of an unnecessary dustup (although the breathless media coverage of ENCODE didn't help). What surprises me especially is that people are surprised by junk DNA.

Even to someone like me who is not an expert, the existence of junk DNA appears perfectly normal. Let me explain. I think that junk DNA shouldn’t shock us at all if we accept the standard evolutionary picture. The standard evolutionary picture tells us that evolution is messy, incomplete and inefficient. DNA consists of many kinds of sequences. Some sequences have a bonafide biological function in that they are transcribed and then translated into proteins that have a clear physiological role. Then there are sequences which are only transcribed into RNA which doesn’t do anything. There are also sequences which are only bound by DNA-binding proteins (which was one of the definitions of “functional” the ENCODE scientists subscribed to). Finally, there are sequences which don’t do anything at all. Many of these sequences consist of pseudogenes and transposons and are defective and dysfunctional genes from viruses and other genetic flotsam, inserted into our genome through our long, imperfect and promiscuous genetic history. If we can appreciate that evolution is a flawed, piecemeal, inefficient and patchwork process, we should not be surprised to find this diversity of sequences with varying degrees of function or with no function in our genome.

The reason why most of these useless pieces have not been weeded out is simply because there was no need to. We should remember that evolution does not work toward a best possible outcome, it can only do the best with what it already has. It’s too much of a risk and too much work to get rid of all these defective and non-functional sequences if they aren’t a burden; the work of simply duplicating these sequences is much lesser than that of getting rid of them. Thus the sequences hung around in our long evolutionary history and got passed on. The fact that they may not serve any function at all would be perfectively consistent with a haphazard natural mechanism depending on chance and the tacking on of non-functionality to useful functions simply as extra baggage.

There are two other facts in my view which should make it very easy for us to accept the existence of junk DNA. Consider that the salamander genome is ten times the size of the human genome. Now this implies two possibilities; either salamanders have ten times functional DNA than we do, or that the main difference between us and salamanders is that they have much more junk DNA. Wouldn’t the complexity of salamander anatomy of physiology be vastly different if they really had so much more functional DNA? On the contrary, wouldn’t the relative simplicity of salamanders compared to humans be much more consistent with just varying degrees of junk DNA? Which explanation sounds more plausible?

The third reason for accepting the reality of junk DNA is to simply think about mutational load. Our genomes, as of other organisms, have undergone lots of mutations during evolution. What would be the consequences if 90% of our genome were really functional and had undergone mutations? How would we have survived and flourished with such a high mutation rate? On the other hand, it’s much simpler to understand our survival if we assume that most mutations that happen in our genome happen in junk DNA.

As a summary then, we should be surprised to find someone who says they are surprised by junk DNA. Even someone like me who is not an expert can think of at least three simple reasons to like junk DNA:

1. The understanding that evolution is an inherently messy and inefficient process that often produces junk. This junk may be retained if it’s not causing trouble.

2. The realization that the vast differences in genome sizes are much better explained by junk DNA than by assuming that most DNA is truly functional.

3. The understanding that mutational loads would be prohibitive had most of our DNA not been junk.

Finally as a chemist, let me say that I don’t find the binding of DNA-binding proteins to random, non-functional stretches of DNA surprising at all. That hardly makes these stretches physiologically important. If evolution is messy, chemistry is equally messy. Molecules stick to many other molecules, and not every one of these interactions has to lead to a physiological event. DNA-binding proteins that are designed to bind to specific DNA sequences would be expected to have some affinity for non-specific sequences just by chance; a negatively charged group could interact with a positively charged one, an aromatic ring could insert between DNA base pairs and a greasy side chain might nestle into a pocket by displacing water molecules. It was a pity the authors of ENCODE decided to define biological functionality partly in terms of chemical interactions which may or may not be biologically relevant.

The dustup from the ENCODE findings suggests that scientists continue to find order and purpose in an orderless and purposeless universe which can nonetheless produce structures of great beauty. They would like to find a purpose for everything in nature and are constantly looking for the signal hidden in the noise. Such a quest is consistent with our ingrained sense of pattern recognition and has often led to great discoveries. But the stochastic, contingent, haphazard meanderings of nature mean that sometimes noise is just that, noise. It’s a truth we must accept if we want to understand nature as she really is.

Adapted from a previous post on Scientific American Blogs.

Modular complexity and the problem of reverse engineering the brain

Bell's number calculates the number of connections between
various components of a system and scales exponentially
with those components (Image: Science Magazine).

I have been reading an excellent collection of essays on the brain titled "The Future of the Brain" which contains ruminations on current and future brain research from leading neuroscientists and other researchers like Gary Marcus, George Church and the Moser husband and wife pair who won last year's Nobel prize. Quite a few of the authors are from the Allen Institute for Brain Science in Seattle. In starting this institute, Microsoft co-founder Paul Allen has placed his bets on mapping the brain…or at least the mouse visual cortex for starters. His institute is engaged in charting the sum total of neurons and other working parts of the visual cortex and then mapping their connections. Allen is not alone in doing this; there’s projects like the Connectome at MIT which are trying to do the same thing (and the project’s leader Sebastian Seung has written a readable book about it).

Now we have heard prognostications about mapping and reverse engineering brains from more eccentric sources before, but fortunately Allen is one of those who does not believe that the singularity is around the corner. He also seems to have entrusted his vision to sane minds. His institute’s chief science officer is Christof Koch, former professor at Caltech, longtime collaborator of the late Francis Crick and self-proclaimed “romantic reductionist” who started at the institute earlier this year. Koch has written one of the articles in the essay collection. His article and the book in general reminded me of a very interesting perspective that he penned in Science last year which points out the staggering challenge of understanding the connections between all the components of the brain; the “neural interactome” if you will. The article is worth reading if you want to get an idea of how even simple numerical arguments illuminate the sheer magnitude of mapping out the neurons, cells, proteins and connections that make up the wonder that’s the human brain.

Koch starts by pointing out that calculating the interactions between all the components in the brain is not the same as computing the interactions between, say, all atoms of an ideal gas since unlike a gas, the interactions are between different kinds of entities and are therefore not identical. Instead, he proposes, we have to use something called Bell’s number B_n which reminds me of the partitions that I learnt about when I was sleepwalking through set theory in college. Briefly for n objects, B_n refers to the number of combinations (doubles, triples, quadruples etc.) that can be formed. Thus, when n=3 B_n is 5. Not surprisingly, B_n scales exponentially with n and Koch points out that B₁₀ is already 115,975. If we think of a typical presynaptic terminal with its 1000 proteins or so, B_n starts giving us serious heartburn. For something like the visual cortex where n= 2 million B_n would be inconceivable, and it's futile to even start thinking about what the number would be for the entire brain. Koch then uses a simple calculation based on Moore’s Law in trying to estimate the time needed for “sequencing” these interactions. For n = 2 million the time needed would be of the order of 10 million years. And as the graph on top demonstrates, for more than 10⁵ components or so the amount of time spirals out of hand at warp speed.

This considers only the 2 million neurons in the visual cortex; it doesn’t even consider the proteins and cells which might interact with the neurons on an individual basis. In addition, at this point we are not even really aware of how neuronal types there are in the brain: neurons are not all identical like indistinguishable electrons. What makes the picture even more complicated that these types may be malleable so that sometimes a single neuron can be of one type while at other types it can team up with other neurons to form a unit that is of a different type. This multilayered, fluid hierarchy rapidly reveals the outlines of what Paul Allen has called the “complexity brake”: he described this in the same article that was cogently critical of Ray Kurzweil's singularity. And the neural complexity brake that Koch is talking about seems poised to make an asteroid-sized impact on our dreams.

So are we doomed in trying to understand the brain, consciousness and the whole works? Not necessarily, argues Koch. He gives the example of electronic circuits where individual components are grouped separately into modules. If you bunch a number of interacting entities together and form a separate module, then the complexity of the problem reduces since you now have to only calculate interactions between modules. The key question then is, is the brain modular, and how many modules does it present? Commonsense would have us think it is modular, but it is far from clear how we can exactly define the modules. We would also need a sense of the minimal number of modules to calculate interactions between them. This work is going to need a long time (hopefully not as long as that for B_{2 million}) and I don’t think we are going to have an exhaustive list any time soon, especially since these are going to be composed of different kinds of components and not just one kind. But it's quite clear that whataver the nature of these modules, delineating their particulars would go a long way in making the problem more manageable.

Any attempt to define these modules are going to run into problems of emergent complexity that I have occasionally written about. Two neurons plus one protein might be different from two neurons plus two proteins in unanticipated ways. Also if we are thinking about forward and reverse neural pathways, I would hazard a guess that one neuron plus one neuron in one direction may even be different from the same interaction in the reverse direction. Then there’s the more obvious problem of dynamics. The brain is not a static entity and its interactions would reasonably be expected to change over time. This might interpose a formidable new barrier in brain mapping, since it may mean that whatever modules are defined may not even be the same during every time slice. A fluid landscape of complex modules whose very identity changes every single moment could well be a neuroscientist’s nightmare. In addition, the amount of data that captures such neural dynamics would be staggering since even a millimeter sized volume of rat visual tissue requires a few terabytes of data to store all its intricacies. However, the data storage problem pales in comparison to the data interpretation problem.

Nevertheless this goal of mapping modules seems far more attainable in principle than calculating every individual interaction, and that’s probably the reason Koch left Caltech to join the Allen Institute in spite of the pessimistic calculation above. The value of modular approaches goes beyond neuroscience though; similar thinking may provide insights into other areas of biology, such as the interaction of genes with proteins and of proteins with drugs. As an amusing analogy, this kind of analysis reminds me of trying to understand the interactions between different components in a stew; we have to appreciate how the salt interacts with the pepper and how the pepper interacts with the broth and how the three of them combined interact with the chicken. Could the salt and broth be considered a single module?

If we can ever get a sense of the modular structure of the brain, we may have at least a fighting chance to map out the whole neural interactome. I am not holding my breath too hard, but my ears will be wide open since this is definitely going to be one of the most exciting areas of science around.

Adapted from a previous post on Scientific American Blogs.

Precision medicine is not precision engineering

From the NYT, a plea for not getting carried away with the vision of 'moonshot medicine' as precision engineering. The op-ed is written by Michael Joyner, a doctor at the Mayo Clinic and takes issue with Obama's precision medicine initiative which he apparently underlined in his state of the union address earlier this month. Obama's proposal is nothing new and has been echoed by proponents since the beginnings of the Human Genome Project. The idea of identifying specific genetic variants in patients and then targeting them with specific drugs seems logical enough. But as the article indicates, we have clearly spoken too soon:

The basic idea behind it is that we each have genetic variants that put us at increased or decreased risk of getting various diseases, or that make us more or less responsive to specific treatments. If we can read someone’s genetic code, then we should be able to provide him or her with more effective therapeutic and preventive strategies. 

But for most common diseases, hundreds of genetic risk variants with small effects have been identified, and it is hard to develop a clear picture of who is really at risk for what. This was actually one of the major and unexpected findings of the Human Genome Project. In the 1990s and early 2000s, it was thought that a few genetic variants would be found to account for a lot of disease risk. But for widespread diseases like diabetes, heart disease and most cancers, no clear genetic story has emerged for a vast majority of cases.

Then there's the psychological aspect:

The push toward precision medicine could also lead to unintended consequences based on how humans respond to perceptions of risk. There is evidence that if people believe they are less at risk for a given disease, they feel excessively protected and their behavior gets worse, putting them at increased risk. Likewise, those who feel they are at greater risk, even if the increased risk is small, might become fatalistic, making their behavior worse as well. Then there are the worriers, who might embark on a course of excessive tests and biopsies “just in case.” In a medical system already marked by the overuse of diagnostic tests and procedures, this could lead to even more wasteful spending.

And finally, the whole idea that biology can be thought of as a linear engineering systems is fraught with flaws and uncertainty (it's also not everyday that you hear the word 'omertà'):

Given the general omertà about researchers’ criticizing funding initiatives, you probably won’t hear too many objections from the research community about President Obama’s plan for precision medicine. But I am deeply skeptical. Like most “moonshot” medical research initiatives, precision medicine is likely to fall short of expectations. Medical problems and their underlying biology are not linear engineering exercises, and solving them is more than a matter of vision, money and will.

This is not the first time that complex medical endeavors have been made to sound much simpler than what they are. The breathless initial promise of the ability of gene sequencing to directly lead to new drugs has not been somewhat dampened, but other efforts still echo this promise, most notably the Brain Initiative announced with much fanfare a few years ago. The goals of the initiative are laudable and some of the technologies that are being envisaged are fascinating, but we are still light years away in any incarnation whatever to "mapping the brain", let alone using that information for direct therapeutic intervention.

Unlike precision engineering where we are dealing with moving parts whose behavior is a matter of well-understood physical principles, the emergent chaos of biological systems is a very different matter. We should certainly not stop trying to conquer this frontier but we should also make sure everyone knows how far the horizons are. Biology is not physics.

"Clueless machines being ceded authority far beyond their competence"

Edge.org which is well-known for asking big-picture questions and having leading thinkers offer their answers to the questions has a relevant one this year: "What do you think about machines that think?" As usual there are a lot of very interesting responses from a variety of writers, scientists, philosophers and members of the literati. But one that really got my attention was by the philosopher and cognitive scientist Daniel Dennett.

I guess I got drawn in by Dennett’s response because of his targeted dagger-thrust against The Singularity ("It has the earmarks of an urban legend: a certain scientific plausibility ("Well, in principle I guess it's possible!") coupled with a deliciously shudder-inducing punch line ("We'd be ruled by robots!"). 

He then goes on to say this:

These alarm calls distract us from a more pressing problem, an impending disaster that won't need any help from Moore's Law or further breakthroughs in theory to reach its much closer tipping point: after centuries of hard-won understanding of nature that now permits us, for the first time in history, to control many aspects of our destinies, we are on the verge of abdicating this control to artificial agents that can't think, prematurely putting civilization on auto-pilot.

The problem thus is not in trusting truly intelligent machines but in becoming increasingly dependent on unintelligent machines which we believe – or desperately want ourselves to believe – are intelligent. Desperately so because of our dependence on them – Dennett’s examples of GPS and computers for simple arithmetic calculations are good ones. The same could be said of a host of other technologies that are coming online, from Siri to airplane navigation to Watson-like “intelligences” that are likely going to become a routine basis for multifactorial tasks like medical diagnosis. The problem as Dennett points out is that belief in such technologies packs a double whammy – on one hand we have become so dependent on them that we cannot imagine relinquishing their presence in our lives, while on the other we would like to consciously or unconsciously endow them with attributes far superior to what they possess because of this ubiquitous presence.

Thus,

The real danger, then, is not machines that are more intelligent than we are usurping our role as captains of our destinies. The real danger is basically clueless machines being ceded authority far beyond their competence.

One reason I found Dennett’s words compelling was because they reminded me of that structural error in the cloud computing paper which I tweeted about and which Derek concurrently blogged about. In that case the error seems to be a simple problem of transcription but it does not mean that computational algorithms cannot ever pick out chemically and biologically nonsensical structures.

Fortunately unlike engineering and technology, biology and chemistry are still too complex for us to consider ceding authority to machines far beyond their competence. But since computers are inevitably making their way into the field by leaps and bounds this is already a danger which we should be well aware of. Whether you are being seduced into thinking that a protein’s motions as deduced by molecular dynamics actually correspond to real motions in the human body or whether you think you will be able to plumb startling new correlations between molecular properties and biological effects using the latest Big Data and machine learning techniques, unimpeded belief in the illusion of machine intelligence can be only a few typed commands away. Just like in case of Siri and Watson, MD and machine learning can illuminate. And just like Siri and Watson they can even more easily mislead. 

So take a look at whether that molecule contains a vinyl alcohol the next time you grab a result from the computer screen and put it into the clinic. Otherwise you might end up ceding more than just your authority to the age of the machines.

What if the Manhattan Project had been like an Alzheimer's disease drug discovery project

The vast K-25 gaseous diffusion plant at Oak Ridge - an
engineering endeavor (Image: Nuclear Secrecy Blog)

Every once in a while you will find someone comparing a major scientific or technological challenge to the Manhattan Project - among such comparisons would be the Human Genome Project and the Brain Map Initiative. It's also not unheard of for drug discovery being uttered in the same breath as the Manhattan project; for instance administrators and scientists have been calling for new antibiotic discovery to be placed on the same footing as the wartime quest to build the first atomic bomb.

To be honest, most of these comparisons obfuscate more than they instruct. It's not that they are entirely invalid, but their core kernels of truth significantly differ.

To see why, let's compare a typically challenging, novel drug discovery project like finding a cure or mitigating therapy for Alzheimer's disease to designing Fat Man or Little Boy. The mandate for the Manhattan Project was, "produce a practical military weapon that works by harnessing energy from nuclear fission of uranium or plutonium". The mandate for Alzheimer's disease would be "discover and develop a small molecule that mitigates or cures the symptoms of Alzheimer's disease and that is potent, safe, enters and exits the body in a reasonable period of time and causes minimal side effects".

The first thing to notice is the difference between the word "produce" and "discover". Manhattan was not a discovery project, and the word "produce" in fact appeared in the first line of 'The Los Alamos Primer', the indoctrination lectures given by physicist Robert Serber at the beginning of the adventure. Production is more akin to engineering than science, so that word sets the tone for the entire project. That is not to say that Manhattan did not involve science and scientists - of course it did. But the key thing to realize is that the basic discovery part of the science had been done between 1938 and 1942. This part was symbolized by four milestones: the discovery of fission in December 1938 by Hahn and Strassmann, the first 'proof of principle' calculations by Frisch and Peierls in March 1940 indicating the feasibility of a fission weapon, the working out of the actual mechanism and effects of a bomb in the summer of 1942 by Oppenheimer and his associates at Berkeley and the successful initiation of a nuclear chain reaction by Fermi and his associates in Chicago in December 1942. By the time the project started in March 1943, the atomic constitution of matter was thus firmly mapped and the elementary particles which were required to harness fission were all discovered and their properties charted as either 'known knowns' or 'known unknowns'.

The major challenges associated with the project were thus not discovery challenges. It was fully known by the beginning of the project that if you suddenly bring a sufficiently large lump of highly purified uranium-235 together you would cause a very big bang. The key words there were 'large', 'purified' and 'suddenly' and these words really signaled the enormous engineering challenges. To produce a large and purified lump of uranium or plutonium would take vast chemical and engineering complexes in Oak Ridge, TN and Hanford, WA which employed hundreds of thousands of workers and whose use of resources like electricity and copper would rival the size of the US automobile industry. That was really mostly engineering, the unimaginably strenuous application of man and machine in separating two isotopes from each other and in creating a novel element.

The 'sudden' part of the challenge was equally important and again heavily steeped in engineering. For the uranium bomb this was not a big issue since a heavy modified gun would do the job. The really novel find - and this would be classified as a discovery, albeit of an applied kind - was the mechanism of a plutonium weapon. In the summer of 1944 it was realized for good that given the spontaneous rate of fission of Pu-240 which was contaminating the samples of Pu-239 produced in the Hanford reactor, a gun-type weapon that would work for uranium would simply cause an equivalent plutonium weapon to pre-detonate. The solution circumventing this problem - implosion using shaped 'lenses' - was probably the most novel discovery/invention to come out of the Manhattan Project. The novelty of this solution was why they tested the plutonium bomb in July 1945 but not the uranium bomb. In fact implosion could be considered to be the one truly novel 'secret' to come out of the project.

But again, putting implosion into practice was almost completely engineering. Brilliant scientists like John von Neumann did contribute in calculating how you would get a perfectly symmetrical, inward-looking shock wave to compress the plutonium core, but the key challenges were designing and fashioning the explosive lenses - layered arrangements of slow and fast-burning plastic charges that would alternately diverge and then precisely converge a shock wave - that would make implosion possible as well as crafting detonators that would fire simultaneously on the surface of the weapon to send the shock waves in. To that end the metallurgists and chemists at Los Alamos were put to work refining these plastic arrangements from hot molds, shaping them and smoothing out even the tiniest air imperfections in the form of bubbles that would cause the shock waves to deviate from perfect symmetry. Similarly experts in electronics were put to work inventing new timing systems for detonators. This work was all chemistry, chemical engineering, electronics and machine shop. It involved not exalted, Nobel Prize-winning minds but the most practical minds, minds which sometimes lacked even a college degree but which could work wonders with their hands (one of these minds happened to be that of David Greenglass, the spy who shuttled secrets to Klaus Fuchs). In fact this part of the project was so important that much of the details are still classified, and rightly so. Master the design of explosive lenses and you command the explosive consequences of plutonium.

The culmination of all this engineering and science is well known, but the tone for that was set in 1943. No wonder Richard Feynman, when describing the Manhattan Project in his memoirs, called it "not science, mostly engineering". He was right.

Now what if the Manhattan Project had been like a novel drug discovery project for Alzheimer's disease? The physicists working on it would still be in the dark ages. The equivalent of fission in Alzheimer's would be the mechanism(s) that causes it, both on the molecular level and on a more global level. We don't know that yet. The Alzheimer's equivalent of protons, neutron and electrons would be the molecular or epidemiological components that cause the disease. There are scattered clues about them, but we really don't know those yet either. Consider the rogue, misfolded protein Aß (amyloid beta) for instance: ten years ago it was regarded as possibly the major culprit in the disease; now it is regarded simply as something associated with the disease in a major way, but nobody knows exactly how. Every major clinical trial that promises to target interesting mechanisms and components in Alzheimer's has failed miserably over the last few years, which is probably not too surprising if we were in ignorance of the real molecular components and were targeting the wrong mechanisms to begin with. 

And even if we knew the mechanisms and the components, the ensuing development of an Alzheimer's drug is no mere engineering challenge. That's because even the most basic processes of drug discovery - things like getting drugs across cell membranes or even getting them to dissolve in an aqueous solution - are too poorly understood to be able to be predicted accurately. Thus, the engineering part of drug discovery is far more tied to a woefully deficient understanding of the science than the engineering part for the Manhattan Project was. The implication of this is that because prediction is largely futile, we have to test tens of thousands of candidates in drug discovery to see what works: What if the Manhattan project needed to build thousands of bomb prototypes to find the one that finally worked? The difference between bomb design and drug design is thus not one of manpower, resources or engineering; it is one of a basic lack of information and deep, dark patches of ignorance. And it arises from the fundamental complexity of biological systems compared to engineering systems.

Thus, if the Manhattan Project were truly an Alzheimer's disease drug discovery project, the physicists working on it would have started not knowing about nuclear fission and not even knowing about protons, neutrons and electrons, let alone about cross sections or plutonium. Here's what their mandate would have looked like then: "Discover what stuff is made up of. Find if any of it can be manipulated to release large amounts of energy. And if you find this, then try to figure out if you can get enough of this special material to make a practical military weapon." In other words, there probably would not have been a mandate to build an atomic bomb in the first place.

However I am ready to take bets on whether this mandate - given to all those brilliant minds in 1942 - would have led to Little Boy or Fat Man by 1945.

Afterthought: So if the Manhattan Project was in fact mostly engineering, it's worthwhile asking why it's associated with science and scientists - and especially physics as opposed to chemistry which was equally important - in the public imagination. I believe the answer in one word is 'myth-making'. Men like Feynman and Oppenheimer are considered so brilliant and fascinating that they have inevitably come to stand in for the whole project.

Other posts on the complexities of biology and the futility of comparing drug discovery with engineering challenges or physics:

1. Why chemistry (and biology) is not physics.
2. Why drug design is like airplane design. And why it isn't.
3. Derek Lowe on what he calls the 'Andy Grove Fallacy': 1, 2.
4. Why it's hard to explain drug discovery to physicists.