Field of Science

The Sheri Sangji accident: The experimental details

Science has just published a summary of the report by California's Division of Occupational Safety and Health about the tragic accident involving Sheri Sangji and tert-butyl lithium. The summary is the most detailed description of the accident that I have seen so far and it makes it clear that there were at least four very significant violations of protocol during the experiment that Sangji was performing. These included:

1. Not wearing a lab coat and other appropriate safety gear.
2. Using a plastic syringe that by definition cannot be oven-baked to remove traces of moisture.
3. Using a syringe with a 2-inch needle that was about an order of magnitude shorter than the recommended length (1-2 ft.). This was a very significant safety breach since it would have required Sangji to tilt the bottle to extract the liquid, thus not only increasing the chances of a spill but also diminishing her general degree of control over the whole procedure.
4. Actually pulling the plunger back rather than let it be pushed by the inert nitrogen pressure from the bottle.

Of these four violations, only the first one can be easily assigned to Sangji herself since lab coats constitute a very general and well-known part of safety equipment. The others are specialized and specific to hazardous substances and their assignment is going to be much more ambiguous. The rub of the matter is going to be in finding out if these violations were the result of inappropriate or insufficient communication by the PI or an oversight on the part of Sangji herself.

From what I can tell, the report seems to lean toward the former possibility. One of the statements I found disturbing was Prof. Harran's admission that he "never discussed with Victim Sangji the risks associated with the tasks she was undertaking". Another important matter which I alluded to in a previous post was the responsibility of senior postdocs and graduate students in the lab, and the report provides a new twist to the issue that I hadn't seen before. Harran says that a postdoc in his group was supposed to train Sangji in the specifics of handling t-BuLi. The postdoc himself admits that he does not have specific recollection of providing "formal" training to Sangji. In addition Harran admits that he never confirmed whether the postdoc had in fact properly instructed Sangji in the use of the hazardous reagent. I would think that the relative apportioning of the blame between Harran and the postdoc is almost certainly going to be a focus during the trial.

None of this is too comforting and it certainly does not sound like it would make it easier for Harran to defend himself. And yet the sad fact of the matter is that this is how many labs around the world probably operate. The PI does not immerse himself in the minutiae of handling specific reagents and leaves it to the postdocs in the group. The postdoc or senior students in turn gingerly step into that notoriously gray area where it becomes difficult to say whether a particular degree of instruction was "sufficient" or not; for instance, was it enough for the postdoc to demonstrate the protocol once? How about twice? How about one actual demonstration followed by two pointed reminders?

These and other questions are almost certainly going to come up during the proceedings and their fuzzy, gray nature is going to make it difficult to assign blame. But the details of the report make it clear that somewhere, sometime, the crucial information undoubtedly slipped through the cracks. And even a clear admission of this fact may make practitioners around the world more vigilant and, one hopes, more humble.

Will quantum physics help us cure Alzheimer's disease?

There's an interesting bit of writing out in the journal ChemMedChem by Jean-Louis Kraus, a medicinal chemist in France who has worked on drug discovery for Alzheimer's disease. The article is essentially a summary of Kraus's pessimistic outlook towards current therapies and approaches addressing Alzheimer's disease. Kraus has worked for a long time in medicinal chemistry and his words reflect experience and not just opinion. The article starts off with some well-founded skepticism but ends up sounding...let's say a little questionable.

For the most part I agree with Kraus that Alzheimer's research in the last decade has seen one disappointment after another and that we are still largely groping in the dark. He refers to the fact that all the concerted research into the disease and the billions of dollars in funding have resulted in only a handful of drugs and a handful of protein targets. None of the drugs even partially repair the damage and none of the proteins have been shown to be decisive as targets for treatment. Beta-secretase, gamma-secretase, the NMDA receptor, acetylcholine esterase; all of these have seen their day in the sun, followed by a disappointing set of results usurping them from front stage. That doesn't mean that none of them are important, it's just that targeting them with therapies seems to have no direct causal connection with treating the disease.

But what is even more disappointing is that the basic hypothesis driving the field - the amyloid hypothesis - has now been seriously questioned. A series of high-profile clinical trial failures have sent researchers scurrying back to their benches and while amyloid almost certainly has an important connection with the disease, it's now not clear at all whether actually targeting the infamous protein aggregates will bring any benefits. What seemed like a promising and rather direct direction of research has devolved into a scientific mess that will need at least a few years to be sorted out. As far as Alzheimer's disease goes, we are still fighting with sticks and stones.

As Kraus says, much of this failure eventually is a failure to understand the basic biology of the disease, which in turn entails a failure to understand the brain on a more general level, including mechanisms of memory generation and storage. Even now, much of drug discovery fails because of ignorance of the detailed biology of the disease and its perturbation by small molecules. What is regarded as a sound mechanistic hypothesis is often thwarted by the complex realities of signal transduction. With Alzheimer's disease we seem to have been biting off more than we could chew, and we need to keep untangling the complex interplay of amyloid, the protein tau, the secretases and a multitude of other biochemical components before we can truly start developing therapies. Thus it's inevitable and essential that in addition to chemists and biologists, we will need crucial input from neuroscientists to target the disease.

But will we also need quantum physicists? Kraus's thoughts on the relevance of subatomic science to AD left me slightly nonplussed. He says that:

The theories behind black holes generally suggest that subatomic particles (electrons, protons, neutrons) are themselves black holes, in which time expands in the opposite direction of our proper (perceived) time. Huge amounts of information could be stored by the spin number of photons present in these particle black holes. Could it be possible that the organization of brain matter, in terms of the properties of subatomic particles (quantum mechanics), confers on brain matter the capacities of memory and cognition, and that these phenomena are not encountered in other types of matter structure in the human body?

Come again? I was not familiar with electrons, protons and neutrons being black holes. But even if they are, I fail to see their direct relevance to understanding memory and cognition. Sure, it's a trivial fact that it's a very specific organization of subatomic particles that leads to a brain rather than to a liver or a chair. But the real action all takes place at the level of aggregates of these particles which we call molecules. I get the feeing that Kraus is indulging in a classic reductionist fallacy here. While subatomic particles do constitute the brain, understanding the brain can only come at a higher level, that of rather old-fashioned physics and chemistry involving ionic currents and neurotransmitters.

But Kraus finds a valuable place for quantum physicists in the war on neurodegenerative disorders:

To me it has become mandatory to create an AD scientific community that includes not only medicinal chemists, pharmacologists, biologists, and medical doctors, but also quantum physicists, in order to understand how aging alters the intimate structure of brain matter, where memory and cognition are located, with the hope of finding new AD treatment research orientations.

To me this sounds suspiciously like Roger Penrose's argument in his rather startling book "Shadows of the Mind" in which he postulated a relationship between wavefunction superposition in quantum mechanics and the growth and shrinkage of microtubules as significantly contributing to consciousness. Even a cursory look at that argument raised serious doubts about the relevance of quantum behavior in microtubules and more formal analysis seemed to confirm these doubts. I am not saying that physicists won't be a valuable asset on a drug discovery team, it's just that they are probably not going to use the tools of quantum gravity to map out cognitive pathways anytime soon.

Somewhat ironically, Kraus ends his piece by extolling the role of a systems biology approach in addressing a problem as complex as Alzheimer's disease. With this I wholeheartedly agree, but systems biology is the opposite of reductionism, where new emergent phenomena provide causal explanations that cannot be reduced to the laws underlying their substrates. We do need a suite of analytical tools operating at various hierarchical levels to address the issue, but given enough time and smart people, we should be able to do the job using standard chemistry and biology, albeit at a more sophisticated level. No fancy biophoton entanglement may be necessary.

Kraus, J. (2011). Why as a Medicinal Chemist I Am Not Optimistic about the Possibility of Finding, in a Reasonable Timeframe, Small-Molecule Drugs Capable of Curing the Evolution of Alzheimer’s Disease ChemMedChem DOI: 10.1002/cmdc.201100431

Introverts, extroverts and modern science

Over at "In the Pipeline", Derek has a post that indirectly asks the following question; all other factors being the same, is modern scientific research more conducive toward introverts or extroverts?

The post was inspired by an Op-Ed in today's New York Times by Susan Cain, a social scientist who has an interesting book arguing that modern professional life's bias toward extroverted behavior may be stifling the kind of creativity engendered by introverts. Whether at home or at work, we are expected to constantly collaborate, interact, email and reach out and perhaps all this is turning a little obsessive. I haven't read the book yet but the author is suggesting that much of the modern workplace not only encourages but often demands constant collaborations, meetings (both offline and online) and social networking in ways that are stacked against quiet introverts whose valuable skills may be lost in the din.

It may be worth noting first that both introverts and extroverts have made monumental contributions to scientific research, although there could be an argument for certain types of research favoring one or the other kind of personality. If most of your research involves working out dense mathematical theorems or tracing the life histories of ants in the Amazonian rainforest for that matter, an introverted personality that isn't too fond of social interaction may be best for you. On the other hand, if your research involves tracking consumer behavior in supermarkets or evaluating the effects of political bias on brains using fMRI, you may be best off being an extrovert who likes to interact with other social creatures.

Notwithstanding this tailoring of certain kinds of personalities for certain kinds of research, it's pretty clear that there are lots of exceptions and that people with diverse temperaments have flourished in scientific research. Henry Cavendish who would probably have received two or three Nobel Prizes had his discoveries been made in recent times was pathologically shy and would fire his maid if she stepped in the same room with him. As a contrast, among modern scientists, Niels Bohr was famous for being someone who thought best by talking to people. In the words of his protege, the physicist John Wheeler, talking physics with Bohr was like playing a one-man tennis match, with the other person serving as a wall to relentlessly bounce off ideas. Bohr would often take his students and collaborators for long walks, sometimes for circular walks around the building with the intent of hammering out his thoughts. The mumbling and the perpetual refinements and revisions of his sentence constructions made the experience a little harrowing for the listener. As Bohr and his victim went round and round, ideas would spin off. In sharp contrast to Bohr, Paul Dirac was equally famous for being taciturn to an almost pathological degree, mostly remaining silent even when questions were asked. Another example is Fred Sanger, the two-time Nobel Prize winner who is one of the most self-effacing and introverted scientists that you can find. In his case his inward-looking personality led to a career-long life in the lab doing technical work with scant regard for interviews, publicity or even scientific writing. Sanger who faithfully retired at the relevant age and now tends roses in his garden says that he has always been good at doing, listening and talking, in that order.

But as far as the trend of modern research and the current emphasis on open science goes, James Watson and Francis Crick's collaboration may be a better role model than Bohr's peripatetic thinking-out-loud style. The two hit it off almost instantly at the Cavendish Laboratories. Watson was irreverent and not afraid to walk up to anyone and ask any question that came to him. Crick was notoriously talkative and loud to the point of being irritating. The two were ideally suited to constantly bounce ideas off each other and rattle off their latest brainwave without any thought of politeness or social etiquette. But the real reason their collaboration serves as an inspiration for modern research is because of their completely open style of approaching the DNA problem. The two would ask anybody, learn any technique, build any model, perform any calculation, read any textbook and consult any reference, all without abandon. Any person, printed source, experimental or theoretical technique was meat and drink to them, even if in one infamous case Watson used Rosalind Franklin's x-ray data without her knowledge. This practice of doing whatever it takes to solve a problem is cardinal to today's multidisciplinary scientific research and it will continue to serve us well.

Open science will indeed bring us enormous benefits but it may also inevitably cast out the valuable outliers. The book by Cain seems to argue that there can be such a thing as too much collaboration if it becomes enshrined in scientific practice as an inviolable rule, leading to those who don't fit the mold being ignored or even ostracized. The real problem is with implicitly or explicitly penalizing introverts who would rather work by themselves. I myself am someone who prefers a balance; I don't like to be secluded in a private office (as I was in a previous job) and certainly prefer an open work space with desks and cubicles. But I would also like to work on ideas in solitude both before and after I discuss them with others. If I were to criticize modern collaborative science, it would be in its tendency to convene meetings or telephone conferences to sometimes discuss even trivial matters. It would be in emphasizing a talent for teamwork so much that not perceiving someone as an extroverted team player almost automatically ejects him from the pool of job applicants in an interview. And it would be in criticizing or even bullying up on someone who does not instantly subscribe to the latest technological breakthrough in online collaboration and data sharing.

There's one more drawback which I think is inherent in this era of constantly collaborative science. It offers little opportunity to work for some time without preconceived notions. Sometimes the best ideas come from people who can think outside the box precisely because they are not already exposed to the conventional wisdom in the field. In some ways, the first few days in a new position offer the best chance to do independent thinking that is unencumbered by groupthink. Bringing a new researcher in your organization or group up to speed by instantly steeping him or her in the group's research philosophy may seem like a good idea, but it may deny you the chance to tap into fresh and original insight. It is important in my opinion to let a newcomer in a research organization take some time and offer his own way of thinking to others rather than have him fall in line with your preferred mindset right away. And then there's the currently unanswered question of how much a world full of distractions that include the internet, phones, email and co-workers is affecting productivity.

Part of the reason why Cain's outlook strikes a chord with me is because of the realization that scientific research has always benefited from personalities, working styles and mindsets that are as diverse as the discoveries they make. Science has had room for all of them; the introverts, the extroverts, the evangelists, the prophets, the quiet lab technicians, the soap box enthusiasts and the madmen. All of these have made modern science, and all of them deserve a place at the table, no matter how eager its inexorable march.

Fraud in a glass of wine

One of the biggest stories in biomedical research during the last decade has been the discovery that certain molecules can mimic the effect of what's called caloric restriction, the reduced consumption of calories, either by starvation or by deliberation. Caloric restriction in turn has been linked quite reliably to a slowdown in aging and a general improvement in metabolism in lower animals like, yeast, fruit flies and certain worms. What was particularly alluring was that these effects seemed to be mediated by a single family of genes through proteins called sirtuins. The implication was clear; not only did we have a handle on a significant component of the genetic basis of aging but we could also potentially mimic the effects of anti-aging genes by drugs that targeted sirtuins.

But what really catapulted the story to public attention was the finding that
resveratrol, a molecule found in red wine, might do this. The presence of a (relatively) cheap edible substance, universally consumed, savored and culturally revered that might slow down aging naturally led to unprecedented public attention. The French and Italians might say "I told you so", but suddenly the holy grail of medical science seemed to be within reach. As usual though, the initial euphoria gradually gave way to a more cautious and tempered belief in the benefits of red wine in mitigating the ill effects of age, and indeed in the whole field of caloric restriction itself. The complete story is fascinating and too convoluted to recount here, but the simple fact of the matter is that the biology of aging is much more complex than we imagined and the initial breakthroughs have not been as unambiguous as they seemed. Not surprisingly, ascribing something as complex as aging and its attendant physiological changes to the action of a single family of genes and proteins has turned out to be simplistic at best (as a comparison, even obesity is thought to be caused by dozens of genes with overlapping effects). In addition, anti-aging effects that got the attention of the New York Times turned out to be significant only in "lower" animals and not in mammals. As it stands today, while research on caloric restriction undoubtedly has great potential, many complications need to be ironed out before the initial optimism can be justified. Curiously, much of the high-profile work in the area can be traced in various forms to a single laboratory at MIT. A recent article in Science does a great job detailing the personalities, the findings and the controversies that sprang from this and other laboratories' work; the entire saga seems fit for a Sinclair Lewis novel.

But whatever the scientific status of the field, its high-profile nature and its potentially revolutionary implications promised ample funding for interested researchers, and over the years it has attracted both highly visible as well as lesser known scientists. One of the individuals who waded into resveratrol territory was Dipak Das of the University of Connecticut Medical School. Over the last few years Das published several papers detailing the beneficial effects of resveratrol in possibly preventing or mitigating oxidative damage caused in cardiovascular and neurological diseases. While most of his research has been published in low-impact journals, it seems that Das was on his way to a lucrative research career involving resveratrol and its role in health and disease.

Until now. It seems that somewhere along the road, he started committing fraud on a massive scale, the likes of which haven't been seen in some time in biomedical research. It started when an anonymous tipster tipped off the university about fabrication in some of Das's papers. The university then launched its own probe and formed a review committee. For the past two years the committee has been working in the shadows with the Office of Research Integrity (ORI) and last week they released their findings in a 50-page document. The findings indicate wholesale fraud, manipulation of results and deliberate doctoring of critical data on a shockingly regular basis between at least 2002 to 2009.

Much of the fabrication centers around Western blots, a research tool that's universally used in all aspects of biological research. The Western blot is to biology what spectroscopy is to chemistry and physics. It essentially detects the presence and identity of specific biomolecules, especially proteins. The proteins appear as dark bands on a white background with different lanes representing different proteins. Western blots are absolutely indispensable in confirming the identity of a novel or unknown protein, and I think it's safe to say that most biomedical researchers who have won Nobel Prizes have used these tools in their research. In case of Das, the report has found dozens of Western blots to be grossly manipulated using image manipulation software.

There are two aspects of the report that bear closer scrutiny. One is the sheer number of Western blots found to have been doctored. The committee examined 26 papers and cited no less than 88 figures which appear to be manipulated (there were also several that appeared normal). This is a staggering amount of manipulation and rules out accidental oversight. Das would have to be involved in a conscious, deliberate and extended effort to tamper with so much data. It's quite clear that the magnitude of the manipulation alone points strongly to purposeful fraud.

The second aspect of the report concerns the great difficulty of detecting the fraud. Western blots seem to be notoriously amenable to manipulation; for instance they prominently featured in another recent high-profile case of fraud in India involving a researcher at the National Center for Cell Science (NCCS). In the report on Das's work, single bands of proteins in Western blots have been enlarged and their borders further magnified to show the contrast between the background for that particular band and for others, indicating that the band in question was copied and pasted. Image manipulation software can sometimes produce such artifacts and some of the data appears like it could also have been the result of negligence or sloppy editing, but the number of such instances again rules out merely these possibilities.

But the difficulty of discerning the doctored data also indicates more generally how difficult it can be to detect fraud in science. Even experienced specialists, let alone laymen or researchers from other fields, wouldn't have thought to look for these minute differences which are evident only in retrospect. To demonstrate the process, at one point in the report the committee actually depicts a mock manipulation protocol where bands are edited or appended to other bands from totally different experiments. The subtleties in the data make it clear that even in the future it would be relatively easy for researchers to get away with this kind of manipulation. What these cases point to is the need for automated systems ("counter-software", if you will) that could detect such fine anomalies in submitted Western blots or other presentations and try to separate artifacts from real red flags. In this case of course, the verdict is unanimous. At the end of the report, the committee finds unambiguous evidence of "unequivocal image manipulation, splicing, background erase and duplication" in a vast number of cases resulting in both data fabrication and falsification.

The debacle is ending in ways that such unfortunate scenarios usually end. The university has already begun proceedings to fire Das from his position. It is very likely that he will never be able to do research again, and that's probably the way it should be given the extent of his fraud. Sadly, Das has not made things any easier by accusing university and department officials of racist prejudice. When you have to resort to such allegations in the face of massive evidence detailing your dishonesty, you only make your guilt seem more likely.

Ultimately this episode speaks as much about the culture of scientific research as it does about the transgressions of a particular researcher. We may not know for some time why Das felt like committing fraud on such a massive scale, but I suspect that the high-profile nature of anti-aging research and the funding that such research commands may have had at least something to do with it. In the last few years, resveratrol, caloric restriction and sirtuins have made it into the public discourse about science like few other topics. The possibility of harnessing all this data to solve the ultimate mystery of aging has ensured both sensationalist news items and eager funding agencies wanting to enable the next breakthrough. When you work in such high-profile fields, it is more tempting to fabricate your results to snare more funding. In this particular case, Das's work was deemed to be low-impact and peripheral to the field and so the damage might be negligible, but in someone else's hands it could well be extensive. The case of Jan Schon immediately comes to mind.

The other reason underlying Das's behavior might simply have been the complexities of biology. Das has been at the university since 1984 and got tenure in 1993, so it's curious why he decided to fabricate results in the last decade or so. I am quite ready to believe that his work on the complex effects of resveratrol on disease may have run into roadblocks, prompting him to start making up results that he wanted to see but which he didn't. Most research these days and especially biomedical research is a complex game. In some ways we are trying to bite off more than we can chew. In such cases wishful thinking can dominate, and when expected results don't pan out because of the complexity of the system under consideration, it becomes easier to succumb to desperation and temptation. The resveratrol story may fit into this paradigm, with initial reports suggesting a tantalizingly simple connection between the drugs and aging and more recent reports questioning this connection. Yet again, nature is not just more complex than we imagine, but it is more complex than we can imagine.

The only remedy for avoiding such debacles may be more acute vigilance, self-policing and an honest willingness to accept failures. And some modesty before nature may be in order here.

Other coverage: Derek Lowe (1, 2), San Francisco Chronicle

Image source

What is your favorite deep, elegant or beautiful explanation (in chemistry)?

Over at Edge, they have a survey asking leading scientists, thinkers and writers about what they think is their favorite "elegant, deep or beautiful explanation". This is meant to be a very general question, not even limited to science and includes responses from people as diverse as the economist Richard Thaler, complexity theorist Stuart Kauffman and Stewart Brand (founder of the Whole Earth Catalog). The answers include ideas, explanations, experiments and entities as general and diverse as the scientific method, genes, Pascal's wager, bounded rationality, relativity and the limits of intuition.

The explanations run across the gamut of the sciences and the humanities including physics, biology, economics, neuroscience, politics and business. But conspicuously absent is chemistry, except for a few peripheral references like Charles Simonyi's listing of Besicovitch's theory of atomic forces. And this is in spite of our friend Derek Lowe of "In the Pipeline" being included in this august list. I was gratified to see a chemist being asked for his opinion, and was somewhat disappointed that Derek's favorite explanation was not chemical (his favorite is the rather deceptively simple notion of "freefall"). I of course don't blame Derek for his choice since there is no law dictating that a chemist's favorite explanation should be from chemistry just because he or she is a chemist. My own favorite beautiful explanation is probably Cantor's notion of multiple infinities.

But I did regret the striking omission of chemistry from the list. Sometime back I had a whole post on elegance in chemistry. And I certainly don't want people to think that deep and elegant explanations are limited to physics and biology, because they are not. Chemistry may not boast of profound philosophical explanatory frameworks like the Big Bang or evolution by natural selection. But it makes up for this fact by creating paradigms that directly touch the lives of millions of human beings in ways much more palpable than the Big Bang and evolution. So I thought I would add my own modest thoughts on my favorite deep idea in chemistry.

There's actually a few things at the top of my list; you certainly don't have to think hard to come up with several foundational chemical ideas. But if you really asked for my absolute favorite deep and elegant explanation, it is the shared-electron chemical bond. That's it. Right there is the simple concept that is at the heart of the material world, a concept that if you think about it has had a staggering impact on our quality of life, our relationships with other nations, our notion of prosperity itself. Chemical bonds as manifested in the foundations of modern civilization have certainly contributed as much to life, liberty and the pursuit of happiness as any scientific idea.

The idea itself as formulated by the great Gilbert Newton Lewis and comprehensible to any high-school student is simplicity incarnated; atoms combine into molecules and form a bond when electrons are shared. Everything that comes after the stating of this fact, important as it is, is details. All the quantum chemical wizardry, the thinking-in-orbitals, the great Gaussian simplification, it's after this basic groundwork has been laid. Heitler and London, Pauling, Slater, Mulliken, Pople, all of them made critical contributions to chemical bonding, but they all stood on Lewis's shoulders and built up from his landscape of the shared electron chemical bond.

Given the absolutely foundational role that the chemical bond plays in the thinking of chemists, it may be both ironic and a tad disturbing that chemists still cannot completely agree on the precise definition of every molecular bond out there. But that's not because the basic framework underlying bonding is uncertain. Part of the reason is simply because there is no such thing as "the" chemical bond. The bonding zoo sports a bewildering variety of animals, from the upstanding "normal" chemical bonds in, say the hydrogen or methane molecules, to the (literally) ready-to-snap pressure cooker entities in strained organic compounds, from the wily, shape-shifting bonds between metals and organic compounds to the ephemeral but biologically vital hydrogen bonds. Although the basic theory of the chemical bond is securely in place, it's going to take some time to craft a net wide and yet rigorous enough to snare the unruly and colorful creatures dotting the chemical landscape.

Now physicists may try to appropriate the chemical bond as their own, but they are out of luck. No explanation based purely on physics can truly impart a feel for the sheer diversity of bonds quoted above and their context-specific personalities. Just one bond serves to create a nightmare for purely reductionist approaches to defining chemical bonding- the hydrogen bond. Last year chemists convened at a meeting with the express purpose of tweaking their description of this all-important biological mediator, the glue that holds life together. Several questions were bandied about, but none more important than the very definition of a hydrogen bond. The problem was simple; hydrogen bonds can be weak or strong, sometimes so weak as to strain the definition of a bond, sometimes strong enough to suspiciously qualify as a covalent bond. How much of hydrogen bonding is electrostatic and how much is covalent? Is "bond" even the right term, or would "bridge" be more accurate? How do you define hydrogen bonds to metals? A consensus was finally reached on a new definition, but not even Linus Pauling could say that the definition would hold for all of eternity. Defining a hydrogen bond would give every physicist out there a run for his money. I find the concept of the chemical bond so enticing and elegant partly because even a single kind of bond like the hydrogen bond can hide a richly textured world of possibilities lurking behind its surface.

So there it is, why the concept of the chemical bond is my favorite idea, certainly in chemistry. It is deep because it underlies the making of the material universe, explaining the stuff that everything from crab shells to the Crab Nebula is made of. It is elegant because of the virtually unlimited amount of explanatory power that it hides in a simple statement of definition. And it is beautiful because of the sheer diversity of materials and structures that are created from a simple law of attraction. A lot of the thinkers in the Edge survey quoted evolution as their favorite deep idea. It certainly is beautiful. But Darwin could well have slightly paraphrased his words to apply to Lewis's shared-electron chemical bond:

"There is grandeur in this view of the material world, with its several powers, having been originally breathed into a single bond; and that, whilst this planet has gone cycling on according to the fixed law of gravity, from so simple a bond endless forms most beautiful and most wonderful have been, and are being, evolved."

Molecular modeling: How far can physics take us?

Of all the scientists writing about modeling and simulation in drug discovery in the last decade or so, I have found Anthony Nicholls of OpenEye Scientific Software to be one of the most insightful. Not only has he written important papers emphasizing the role of rigorous statistics in generating and communicating modeling results, but he has also been a relentless proponent of the need for rigorous, unglamorous but essential experimental data to validate modeling protocols. In the phalanx of modelers pointing to a better future for their field, Anthony has been one of the torchbearers. I usually pay close attention when he writes so I think it's worth noting what he has to say in a recent article titled "The character of molecular modeling".

He starts by asking what the real advances in the field have been in the past 25 years and by observing an apparently rather disconcerting fact about modeling and especially structure-based modeling - successes are still mainly anecdotal. He expresses his disappointment while noting that most of the successful results in modeling are still mainly of the "find protein pocket, fill pocket" type. The chief role of the crystallographer, it seems, is to supply pockets that the computational chemist can then fill. The problem according to Anthony is that chemists are not "abstracting principles of wide applicability; they are recognizing domains of expertise".

At this point let me interject and say that while Anthony's gloomy prognosis might be true, it's also true that "find pocket, fill pocket" (or "find pocket, kill pocket" if you are in a hunter-gatherer mood) campaigns are not as straightforward as we think. There can be unexpected effects on both protein conformation and ligand conformation, similar to the "activity cliffs" witnessed by medicinal chemists. Even if the binding orientation of the ligand stays constant upon small changes, the distribution of solution conformations of the modified ligand is likely quite different, leading to differing energetic penalties that the protein has to pay for binding. I am sure I am not alone in saying that small changes in ligand structure leading to changes in binding affinity enforced by ligand strain and conformation are uncomfortably frequent. But there's another dimension to the "find pocket, fill pocket" campaign; it can actually be quite satisfying to suggest changes to a medicinal chemist for filling the pocket that are borne out by further crystallography. Finding pockets may generate anecdotes, but chemistry is a more anecdotal science than say physics, and chemists often revel in these little successes and failures. Chemists are more frogs than eagles.

But the real sticking point for Anthony is not really the anecdotal success of structure-based modeling but the lack of general physics-based principles and laws for doing molecular modeling. Docking is an example. In the last several years there have been many attempts to use physics-based "scoring functions" - essentially ways to sum up different protein-ligand interactions to a number - for calculating the binding affinity of a ligand. Programs for docking have evolved to a stage where ligands can be docked in the correct orientation with a roughly 30% success rate, depending on how similar the docked ligands are to a reference co-crystallized ligand. But the truth of the matter is that we still fail miserably when trying to dock an arbitrary ligand to an arbitrary protein in an arbitrary conformation. And of course, we are light years away from predicting free energies of binding for the general case. There have been cases in which physics in the form of electrostatics and quantum mechanics (more on this later) has significantly accelerated the search for similar molecules, but the promised land still seems far.

Does this failure reflect an absence of general principles of physics for computing protein-ligand interactions? Paraphrasing Rutherford (not Niels Bohr), in the next few decades will we do more physics or simply collect more stamps? Is this concern even warranted? To some extent, yes. It would certainly be very satisfying to have a general explanatory framework, a pool of more or less universal laws that explained the wide variety of protein-ligand complexes as completely as Newton's laws explain the behavior of an astonishingly diverse set of particle interactions in the classical world. Curiously, such a general framework does exist in the form of statistical mechanics and quantum mechanics. In theory, both these disciplines encompass the binding of every single protein to every single drug. So does that mean we can look forward to a time when every modeler can "abstract these principles of wide applicability" and use them to solve the particular case of his or her protein and ligand?

Here is where I part ways with Anthony at least partly. The reason in my mind is not too hard to discern. Think about how far we have come in explaining protein-ligand binding using the rather extensive developments in either quantum or statistical mechanics over the past five decades. The answer is, not as far as we would have liked to. While we have indeed made great advances in understanding the basic thermodynamics of protein-ligand binding, we have not been very successful in incorporating these principles into predictive computational models. Why so? For the same reason that we have not been successful in using physics to explain "all of chemistry", in Paul Dirac's words. Quantum mechanics has been applied to chemistry for fifty years and exponentially increasing computational power has significantly furthered its application, but even now, for most practical systems chemists use a variety of empirical models to understand and predict. That's partly because most real systems are too complex for the direct use of quantum mechanics, and an imperfectly understood protein and ligand immersed in an imperfectly understood solvent certainly belong to this category. It's also because we are still far from calculating things like entropy and being able to model the differential behavior of water at interfaces and in the bulk.

But even more importantly, physics may not solve our problems because chemists need to abstract general principles at the level of chemistry to ply their trade. Thus, in expressing doubts about the utility of general physics-based principles, I am appealing to the strong sense of non-reductionism that permeates chemistry and separates it from physics. The same principle applies to biology and I have written about this often. Principles drawn from physics have always been very useful in gaining insights into molecular interactions and they will continue to be an essential part of the mix. But unlike Anthony, I see a far smaller role that pure physics can truly make in enabling a general, practical predictive approach to modeling that's "chemical" enough to be widely used by chemists.

So are there cases in which physics can make a contribution? Here I actually do agree with Anthony when he mentions two areas where physics really promises to have a substantial impact, both conceptually and practically. The first is crystal structure prediction for organic molecules which is a notoriously fickle problem (a measure of the difficulty can be gleaned by the fact that even the simple benzene can crystallize in more than 30 different geometries), essentially one of being able to predict fine energy differences between almost equienergetic arrangements. Yet I see this problem as one of the more reductionist problems in chemistry, and as Anthony notes, it is conceivable that it will yield to physics-based approaches in the near future.

The other problem is one of the holy grails of chemistry and biology - protein structure prediction. In various guises, the last few years have seen a startlingly impressive set of cases where protein structures of small and (some) medium-sized proteins were predicted with atomic level accuracy. Protein structure prediction has to overcome the twin challenges of sampling and energy estimation that are a mainstay of almost every other modeling method. In this case Anthony thinks that we will have to get the physics right to address this issue.

But we have to be careful to distinguish between two cases here. The first case is where we get the right structure even if we have no idea how we got there. This is the field of empirical (non-physics based) protein fold prediction and the biggest success in this area has been the ROSETTA suite of programs. ROSETTA has definitely turned heads within the community by its ability to generate accurate structures for hundreds of proteins, but the big drawback of the approach is that it only generates the end result. Curiously Anthony does not mention ROSETTA, but I am also surprised that he does not mention in detail another significant development that does fit into the physics-based paradigm. This is the molecular dynamics
approach developed by David Shaw, Vijay Pande and others. Unlike ROSETTA, MD can actually shed light on the process leading to a correct structure, although the details of the process are subject to errors, most notably in the force fields that underlie the simulation. It's quite clear that with all their limitations, ROSETTA and MD have been the biggest contributors to successful protein folding simulations over the last decade.

And yet as Anthony rightly says, their success seems almost like a miracle. This becomes clear when we realize that even now we have trouble predicting something as simple as the solvation energy of a simple organic molecule or the interaction energy of two simple molecules using even sophisticated quantum mechanics calculations. If our ability to predict even such simple scenarios is dismal, how on earth are we getting the structures of all those complex proteins right? The answer deserves as much scrutiny as the solution to these problems, scrutiny that is severely lacking. Anthony's answer (and mine) is "cancellation of errors along with a need to calculate only relative, not absolute, energies" (it's well known that force fields are virtually worthless for the calculation of absolute energies). It still strains my mind to think that these two factors could contribute to so many successful predictions published in the likes of Nature and Science. Cancellation of errors was partly made famous by Enrico Fermi. If that's really what's happening in all these cases, then the entire field needs to start celebrating Fermi as their guardian angel.

Ultimately, there is no doubt that advances will continue to be made with increasing computational firepower, but the foundations of the field will stay brittle unless these fundamental issues are addressed. Anthony ends with something he has been doing for a long time now - appealing to experimentalists, industry and government to contribute a small part of their funds to the kind of basic experiments that can further the field of modeling. This especially involves experiments that can refute an idea, a philosophy that has been dominant in the practice of science since its modern conception but one which seems to be unusually neglected in drug discovery because of the emphasis on positive data gathering. Science has always progressed by the testing of ideas that have no immediate practical bearing, except that they perform the invaluable function of making future scientific research worthwhile. It would be fundamentally unscientific if such ideas are not supported. Anthony puts it well:

"The simple commitment to spend a small percentage of the science budget at the NIH or at pharmaceutical companies on nontranslational work, providing support for the small cabals of scientists actually interested in making fundamental progress would be enormous. Reestablishing the contact between theorists and experimentalists, the publishing of high quality data, conferences devoted to the actual testing of ideas—in 25 years we might hope molecular modeling could become a real scientific discipline."