Skip to main content

We Don't Understand Life

Otherwise, we'd be able to build it from scratch.


One way to show that you understand how something works is to build it. Circuit boards, computers, skyscrapers, and internal combustion engines demonstrate that we understand electronics, computation, structural engineering, and thermodynamics. If we didn't understand gravity and Newton's laws of mechanics, NASA's Curiosity rover would have never even made it to the vicinity of Mars, and without knowledge of the workings of chemical reactions, the plastics, fabrics, films, and other synthetic materials we depend on wouldn't exist.

If building something is a test of our understanding, then clearly we have failed to understand something important about living things, because we have utterly failed to build life from scratch. Will we ever do it?

One of science's big, unsolved mysteries is how life first arose from non-living chemistry on Earth, four billion years ago. Despite decades of effort, we've never been able to evolve living things from pure chemical components in the lab. This failure reflects the fact that we don't understand something crucial: We don't understand the environmental conditions and chemical ingredients that gave rise to the primitive, self-replicating chemical systems that were the direct predecessors of Earth's first living organisms. Not only have we failed to understand how life came to be on Earth, but without this understanding we also don't know how likely it is that any of the billions of habitable planets in the Milky Way are actually inhabited.

Biologists are like someone who knows how to work a 3-D printer but can't design new digital templates for it.

Researchers aren't completely clueless about the origins of life. Two necessary features of life are metabolism (the ability to build your own components, including a new copy of yourself, from the raw materials around you) and inheritance (the ability to store and pass on the operating instructions to the next generation). One molecule that researchers have discovered that can take care of both jobs is ribonucleic acid, or RNA, an essential component of all life today. RNA has the ability to both encode genetic material, and to act as an enzyme that builds biological parts.

The ability of RNA to be both genetic material and an enzyme led scientists to propose the "RNA World" hypothesis, which is the idea that very early in the history of life, primitive living systems were made largely of RNA. In the RNA world, so the theory goes, RNA did the two jobs that are now done separately by proteins (metabolism), and DNA (inheritance).

Building on these ideas, scientists are creating self-reproducing systems that come very close to the boundary between non-living chemistry and life. For years, scientists have tried to make RNA replicate itself in a test tube, with some limited success. Researchers have had even more luck with other types of molecules that may have been present in the first living systems. A team led by Jack Szostak at the Massachusetts General Hospital recently showed that a molecule very similar to RNA can rapidly and accurately copy itself much better than RNA. By putting a self-copying genetic system like this into simple, lipid bubbles, researchers can temporarily produce something that looks a lot like a primitive cell.

RATHER THAN EVOLVE LIFE from chemistry in a test tube, an alternate way to build life from scratch is to design it on a computer. This is the goal of a team of scientists at the J. Craig Venter Institute, who in 2010 reported the results of the world's first artificial genome transplant. These scientists began with a computer file of the DNA sequence of a small bacterium, Mycoplasma mycoides. Using that computer file as a blueprint, the researchers chemically synthesized the DNA to create an artificial genome. They then took the DNA and transplanted it into the cell of a different species of bacteria, and thereby transformed one species into another.

While this was a clever, technically stunning feat, they didn't quite create life from scratch, because they started out with the code of the existing, natural genome of Mycoplasma mycoides. In fact, one single mistake in their copy of that genome caused the experiments to repeatedly fail, until the researchers corrected the error. What these scientists (or anyone else) cannot do is sit down at the computer and design a brand new genome, freshly coded from the ground up. Biologists are like someone who knows how to work a 3-D printer but can't design new digital templates for it.

To figure out what it takes to build a basic, functioning genome, researchers are working backward by stripping away parts to find the “minimal genome,” the bare bones set of genes that will get you a working organism. The organism with the fewest genes so far is Mycoplasma genitalium, a bacteria that lives in the human urinary tract and has only 485 genes—and even 100 of those seem to be dispensable. (For comparison, bacteria like E. coli have about 4,500 genes, and humans have more than 20,000.)

Why search for a minimal genome? Because this minimal genome could serve as a "chassis," a basic design platform that could be modified to create designer cells with the ability to make our fuels, drugs, and plastics in a way that is more resource efficient and less environmentally disruptive.

SO WHAT'S STOPPING US from successfully building life from scratch? When it comes to evolving life from chemistry in a test tube, the answer probably is that life will arise under a few very specific chemical conditions, out of an enormous number of possibilities. Scientists can only find those conditions by rationally guided trial and error. But after a first success, the necessary features of other conditions that can give rise to life will be more obvious, and we'll better understand not only how life first arose on Earth, but also how much life we can expect to exist in our galaxy.

Intelligently designing entire new genomes, without simply copying nature's successes, may be a more difficult challenge, because we don't even know how to design the parts. We're terrible at designing new genes (as opposed to simply hacking existing ones). And even if we could design them well, getting hundreds of newly designed genes to seamlessly work together in a functioning system is well beyond our current capabilities. Designing a brand new genome with our current knowledge is like trying to build a Boeing 787 without circuit diagrams, a functioning machine shop, or even a basic understanding of aerodynamics. In the near future, designer genomes will be variations on nature's handiwork.

If we ever do create life from scratch, what does that mean for us? So much of who we are is now explained by a series of physical causes: We're made of molecules that are not unique to life. We develop from a single cell into an adult by following a self-sustaining genetic program that requires no outside, intervening, vital force to guide the physical process. And we are not only the descendants of small-brained savannah primates, but we also bear the clear genetic signature of our primal, single-celled ancestors who floated around in Earth's warm oceans more than three billion years ago. While there are many gaps to be filled in these explanations, three very deep gaps in science's explanatory scheme of the physical causes that created us remain: the origins of consciousness, the origins of life from chemistry, and the ultimate origins of the universe. We'll probably never be able to build a universe, but some day we will demonstrate our understanding of life and consciousness by creating them.