My colleague George DeMartino at the University of Texas Southwestern Medical Center at Dallas told me a humorous story that indicates the devolved state of our beloved field of biochemistry. George uses biochemical approaches to study the proteasome, including standard methods of protein purification. While interviewing a prospective postdoctoral candidate, George explained how he routinely purifies the proteasome, starting with bovine blood obtained from a local slaughterhouse. As the interview progressed, the applicant lauded George on his ability to study the purified macromolecular machine and stumbled into the telling question of the interview: “That’s really cool, but how did you get the His tag into the cow?”
It is the unfortunate case that few of our trainees have any clue about ammonium sulfate cuts, differential centrifugation or column chromatography. Young scientists instead think affinity-precipitation of a tagged, overexpressed protein constitutes “biochemistry.”
Few would dispute that our field is out of fashion. What is the in-vogue science that obscures biochemistry like a full eclipse of the sun? Topping the list is the big data science evolving from genomics.
The ever-expanding iterations of -omics research offer limitless access to data. The challenge of gathering fresh data used to be difficult; now it’s a piece of cake. I venture to guess that the amount of data gathered and published by the ENCODE consortium last year might, in aggregate, constitute a larger amount of data than what has been accumulated in the entire history of the field of biochemistry.
There are two wonderful things about the gathering of huge data sets. First, it is can’t-fail science. If I tell a trainee to immunopurify fragmented chromatin with an antibody to one of our transcription factors and then have our genomics core sequence the precipitated DNA, the experiment will work every time. What a deal it is to carry out fail-safe experiments! Second, the top-tier journals eat this sort of research up as if it were $1,000-per-ounce caviar. Those of us who have stuck with difficult and uncertain biochemical research are viewed as village idiots – how could we be so stupid not to see the light?
My friend Deepak Nijhawan offered a visual correlative of big data science. When we take our kids to a venue offering a variety of arcade games, they gravitate to the game that consists of a claw that can be moved in X, Y and Z dimensions by a joystick. Below the claw lies a carpet of stuffed toys, perhaps including frogs, bunnies, crocodiles and bears. It is so incredibly easy!
Our kids put their quarters in, maneuver the claw and drop it into place to retrieve the exact stuffed animals of their desires. Over time, they learn the hard way that this never works. Indeed, I have never seen it work a single time in my entire life. What a racket – an infinite number of quarters for nothing. Even if the claw wins once in a thousand attempts, the reward is a stuffed animal that may have cost less to produce than the 25-cent price of admission! Just as our kids mindlessly feed the claw, the National Institutes of Health feeds big data science. Time will tell if the investment will pay off.
|A claw crane game in Trouville, France, promises plush unicorns. Courtesy of Wikimedia Commons user Nlan86.
Here is the idea on big science. Once we gather enough of it, really smart people will be able to extract all of the diamonds of biology. Magically, for example, they will be able to use big data to predict correctly that cells have an enzyme that senses intracellular DNA and then triggers the production of cyclic GAMP, which then activates the STING enzyme to mount an innate immune response. When this happens, we won’t need the biochemical skills of James Chen, who painstakingly discovered the aforementioned pathway.
Had we had access to big data science 30 years ago, we would not have needed Tom Cech’s chemical and biochemical acumen to discover catalytic RNA. Geez, would life have been simpler! The magic claw of big data science could have seen all of the discoveries of significance and simply plucked them out of the pile of massive data sets.
OK, enough foolishness. There is a place for everything, including big data science. The point I seek to make in this inaugural essay is the simple prediction that, as we peer into the looking glass of the future, Chen- and Cech-like discoveries abound. Not being a gambler, I will not short the stock of big data in anticipation of the bursting of its bubble. On the other hand, given that the market cap of mechanistic biochemistry may be at an all-time low, I could not be more bullish on our stock.
Time will tell whether big data science is just a Ponzi scheme or will instead dazzle us with magnificent discoveries. If it does, the reductionist, mechanistic approaches now out of fashion may fade into extinction. I trust that readers will see where my money is: We biochemists are down but not out.
Steven McKnight (firstname.lastname@example.org
) is president of the American Society for Biochemistry and Molecular Biology and chairman of the biochemistry department at the University of Texas-Southwestern Medical Center at Dallas.