How a Directionless Path Can Reveal Science’s Most Closely-Guarded Secrets

Discovery can be a messy business. Rarely is there a straight line to new knowledge; instead, hypotheses and interpretations must navigate a roundabout and often directionless path. But rather than an obstacle to wisdom, this is better considered a necessary part of the process. We have already seen several instances in which important discoveries came about when researchers abandoned their original ideas, creating bias-free conditions that fostered new paradigms. Cells, genes, regulation, induction, transcription, stem cells, reprogramming—all were amorphous inklings, closer to guesses than theories before they became bedrocks of biology.

Researchers and writers have proffered various metaphors to describe this ethereal quality of science, but my favorite comes from François Jacob. In his memoir, The Statue Within, Jacob divides scientific inquiry into two categories, which he refers to as “day science” and “night science”:

Day science employs reasoning that meshes like gears and achieves results with the force of certainty. One admires its majestic arrangement as that of a da Vinci painting or a Bach fugue. Conscious of its progress, proud of its past, sure of its future, day science advances in light and glory. Night science, on the other hand, wanders blindly. It hesitates, stumbles, falls back, sweats, wakes with a start. Doubting everything, it feels its way, questions itself, constantly pulls itself together. It is a sort of workshop of the possible, where are elaborated what will become the building materials of science. Where hypotheses take the form of vague presentiments, of hazy sensations. It is impossible to predict whether night science will ever pass to the day condition. When that happens, it happens fortuitously, like a freak. What guides the mind then, is not logic. It is instinct, intuition.

Nature reveals its most closely guarded secrets preferentially to those who can sit with uncertainty.

Uri Alon, a biologist at Israel’s Weizmann Institute, has another name for the space where basic assumptions break down and new concepts can be born—a realm he calls “the cloud.” Scientists who find themselves in the cloud are filled with anxiety, like hikers lost in the woods craving the safety of the trail. It is an untethered feeling, the sense of being in a room with no exit. It is especially unsettling for students, whose preconceptions of laboratory research are grounded in a cycle of hypothesis, experiment, and refined hypothesis. If only it were that simple! Instead, nature reveals its most closely guarded secrets preferentially to those who can sit with uncertainty, while more rushed visitors are likely to miss out.

The importance of meandering as a prerequisite for discovery becomes even more evident if we compare basic science, the type of exploratory research we have mostly focused on, to engineering, with its defined goals and timelines. Bill Kaelin, winner of the 2019 Nobel Prize in Medicine, described the difference succinctly, noting that engineers are accustomed to “using the rules” while basic scientists are more interested in “learning the rules.” At its most fundamental level, engaging in basic science is an effort to learn a new language—one whose very existence is unknown at the outset.

Of course, this type of fundamental research is not without application. The most profound advances in medicine over the past 20–30 years can trace their origins to studies in plants, bacteria, phage, and flies—work that an uninformed observer at the time might have viewed as irrelevant. We need look no further than the COVID-19 vaccines—mRNA preparations that owe their existence to a chain of discovery starting with Jacob’s phage work—to grasp the power of night science.

These distinctions have implications for the biomedical research enterprise, where the thirst for impact, or “deliverables,” draws resources from more open-ended projects that may carry greater long-term (albeit less visible short-term) potential. “Moonshot” operations—vast initiatives involving dozens or hundreds of scientists with a single-minded goal—are based on the engineering premise that money and sweat can fill a knowledge gap. To be clear, there is a place for such undertakings, especially when the goal is well defined. This was the case for the Human Genome Project—the multibillion-dollar venture to determine the complete DNA sequence of human beings. But when the path of discovery is less clear—how to cure cancer or autoimmune disease or how to comprehend human consciousness—night science, despite its slow pace, is where breakthroughs come from.

“Basic scientists are increasingly asked to certify what they would be doing with their third, fourth and fifth years of funding, as though the outcomes of their experiments were already knowable,” Kaelin writes. It is a kind of thinking that neglects the unpredictability of new knowledge and ignores the history of night science as the disruptive force behind discovery.

The intricacies of cellular memory

The great paradox of night science is that it is invisible before its transition to daytime. The closest we can come, prospectively, is to consider a problem that lives at dawn, neither shielded by darkness nor fully illuminated by sunlight. The example I have in mind is epigenetics, a rapidly growing field that can best be understood in the context of two conflicting facts we already encountered.

Fact number one: All cells contain an almost identical complement of DNA. This is the principle of genomic equivalence, which ensures that a cell retains a complete set of genes as it differentiates.

Fact number two: Cells have memory. As the embryo grows and its cells differentiate, the citizens of our cellular societies hold fast to their occupations and pass them along to their children.

A consequence of fact number one is that cells carry excess information. Hepatocytes possess all the genes needed for neuronal function, and leukocytes have all the information needed to produce cartilage. This creates peril for fact number two, for it leaves open the possibility that at any moment a cell might go “off script” and express an inappropriate cellular program. Imagine the bodily chaos that would ensue if our cells started to forget who they were—if liver became lung or brain turned to bone. In principle, such dramatic transformations are possible, as Gurdon and Yamanaka showed us with animal cloning and cellular reprogramming. But under normal circumstances, this does not occur. Cells hold fast to the identities conferred during development and pass those identities along to their progeny.

Night science, despite its slow pace, is where breakthroughs come from.

Cellular phenotype is a function of gene expression—which genes are ON, and which are OFF. Gene expression, in turn, is regulated by DNA-binding proteins, the transcriptional activators or repressors we learned about in Chapter 4 that regulate the production of mRNA. For bacteria and phage, these gene regulators were all that was needed for a cell to adapt to its environment. When Jacques Monod switched the diet of E. coli from glucose to lactose and back again, the organisms happily flipped back and forth; transcription factors alone mediated the response to changing food sources. The microbes had little need for memory.

But in multicellular organisms, where occupational recall is essential, cells had to find a way to remember their identities and bequeath the information to their descendants. The ability to pass differences in gene expression to the next generation, all in the face of a uniform genome, was yet another facet of the One Cell Problem for nature to solve. It did so, quite ingeniously, with a phenomenon known as epigenetics.

In the early 1940s, biologist Conrad Waddington coined the term “epigenetics” to describe a different mode of heredity, one in which information could be passed down between cells in the developing embryo but not between animal generations. Waddington’s proposal preceded the recognition that DNA is the genetic material, and so his ideas lacked molecular footing. But he perceived that developing cells must have a way to transmit information as they divided—information regarding which genes should be ON and which should be OFF. Epigenetics encapsulated the notion that cellular memory is maintained through a new mechanism of heredity, one distinct from the genetics of Mendel, de Vries, and Morgan. In other words, while genetic mechanisms mediated the heritability of animal form during evolution, epigenetic mechanisms mediated the heritability of cellular identity during development.

Waddington provided a visual aid to help biologists understand this concept—a metaphorical “epigenetic landscape” that imagined cells as balls rolling downhill as they differentiated. In the standard “Waddington diagram,” balls at the top of the hill are meant to represent multipotent cells with the potential to generate many types of offspring, while the valleys below represent those different fates. As cells start their downward journey, inductive forces—signals received from neighboring cells—nudge them in one direction or another. A cell that has access to many paths at first (plasticity) would find its ability to end up in other parts of the landscape growing smaller and smaller as it rolled downhill (commitment). Cellular memory, according to Waddington, is a kind of molecular gravity—it prevents cells, and their progeny, from changing their identities in the absence of some extraordinary force of displacement.

_______________________