In 2010, a team of scientists announced that they had created a synthetic living cell. The team, led by Nobel laureate Ham Smith, microbiologist Clyde Hutchison III, and genomics pioneer Craig Venter, fashioned the full genome of a tiny bacterium called Mycoplasma mycoides in their lab, and implanted the DNA into the empty cell of another related microbe. They nicknamed it Synthia. Some news sources claimed that the team had, for the first time, created artificial life. Others noted that they had merely photocopied life, putting an existing genome into a new chassis, like a “hermit crab taking up residence in an abandoned shell.”

But amid the hyperbole and skepticism, the team continued working. “The 2010 paper was basically the control experiment,” says Venter. Their true mission was to create a cell with a minimal genome.

All living things evolved from a common ancestor, so despite our grand variety, we all share genes that are essential for our survival. They’re at the core of our operating systems: the fundamental software without which we would die. Smith, Hutchinson, Venter, and their colleagues wanted to create an organism with just these essential genes—only those it needed to survive, and nothing more. A minimalist microbe. Kondococcus, perhaps.

Why bother? Because they ultimately want to intelligently design new life-forms from scratch—say, bacteria that can manufacture medical drugs, or algae that churn out biofuels. And creation requires understanding. “We had to start with a system where we knew and understood all the components, so that when we added specific ones to it, we could do so in a logical design way,” Venter says. They needed a minimal genome—a vanilla model that they could later kit out with deluxe accessories.

And they’ve done it. Six years after Synthia, they’ve finally unveiled their bare-bones bacterium. And in piecing together its components, they realized that they’re nowhere close to understanding them all. Of the 473 genes in their pared-down cell, 149 are completely unknown. They seem to be essential (and more on what that means later). Many of them have counterparts that are at work in your body right now, probably keeping you alive.

And they’re a total mystery.

“We’ve discovered that we don’t know a third of the basic knowledge of life,” says Venter. “We expected that maybe 5 percent of the genes would be of unknown function. We weren’t ready for 30 percent. I would have lost a very big bet.”

Back in 1996, when the team began their 20-year quest, Arcady Mushegian and Eugene Koonin estimated that the minimal genome consists of around 256 genes. (For comparison, we have 20,000 to 25,000, the well-studied E. coli has 4,000 to 5,000, and the smallest free-living bacterium Mycoplasma genitalium has 525.) Mushegian and Koonin compared M. genitalium’s  minuscule genome with that of another small bacterium and found 256 overlapping genes. Those, they reasoned, “are close to the minimal set that is necessary and sufficient to sustain the existence of a modern-type cell.” Other scientists have since repeated the same exercise and arrived at a similar answer: life depends on a common core of 200 to 300 genes.

“We set out relatively certain that we could design a cell from scratch,” says Venter. “We were so cocky that we even had a contest between ourselves to see who could do it first.”

They started with Synthia’s genome and deleted 440 seemingly disposable genes, leaving just 432. Once again, they synthesized this set and transplanted it into an empty cell. Which promptly died. “Nothing worked,” says Venter. “We couldn’t get a living cell. It became clear that all those earlier studies were missing something fundamental.”

There were two big problems. The first is clear in hindsight: there are many essential genes that no one knew anything about. The second became obvious earlier: many genes are redundant. The team had been judging the worth of genes by hobbling them one by one and seeing if cells still survived. But many pairs of genes understudy for each other: losing one is no big deal but losing both is catastrophic. Put it this way: knock out either engine on a jumbo jet, and it’ll probably still fly; bill them both as “non-essential” and your plane will crash.

Figuring out these redundancies took a lot of time, especially since many of them involved unknown genes. The JCVI team carried out several cycles of designing, building, and testing, restoring important genes that had been unadvisedly removed and removing more that actually were dispensable. After a few years, they got their first viable cell, which they called syn2.0. It has 516 genes, just nine fewer than M.genitalium. “It wasn’t a moment of celebration,” says Venter. “It was more relief. It told us that we hadn’t done something stupid.”

They finally jettisoned 43 more genes to arrive at syn3.0—their “working approximation of a minimal cell,” with just 473 genes.

Are these “essential”? It depends.“Essentiality is based on the environment,” says Venter. “That’s why we talk about a minimal genome, not the minimal genome.” A microbe growing in a hot volcanic spring needs very different genes to one growing in your gut. And many bacteria live only inside the cells of insects and other animals. Within these constant, closeted environments, the microbes can afford to lose genes that they would otherwise need for a free-living existence. The smallest of them, Nasuia, has just 137.

So there’s minimal, and there’s minimal. Syn3.0 represents the former—something close to the bare set of genes needed for independent life. It can live on some basic nutrients, grow at a reasonable pace, and reproduce on its own. “I think it is difficult—near impossible—to define a minimal genome, but this paper takes us awfully close,” adds Farren Isaacs from Yale University. “It’s an impressive tour de force.”

The team’s top priority is now to work out what those 149 unknown genes do. Some look like they create molecules that stick out from the surface of the cell—but to what end? Others look like they shuttle molecules in and out of the cell—but which molecules? “These are key biological functions affecting all of life that we don’t understand,” says Venter. “It’s doubtful that any federal funding organization would give us a grant to study genes of unknown function. It’s not enough to discover that they’re essential for life. It would be considered a fishing expedition. Maybe that’s why we have so many unknowns.”

These question marks don’t bode well for the ultimate goal of designing genomes from first principles, and synthesizing life from scratch. But in the near-term, the same technology and ethos that led to syn3.0 “can be used to construct new cells with desired properties, opening the door to a real synthetic biology era,” says Rosario Gil from the University of Valencia. “This sure represents an outstanding unprecedented step.”

Venter’s group have already started to tweak the genes they do know about. For example, they took an eighth of syn3.0’s genome and reordered the genes within it according to their function, much like defragging a hard drive. (The result was a delightful piece of organization porn that wouldn’t go amiss on Things Organized Neatly.) To their surprise, the rearranged genome still worked—a reassuring result for a project beset by setbacks.

This philosophy of learning basic biology by building stuff is the best bit of the syn3.0 story, says Drew Endy from Stanford University. “Too often in biology we end up with only data, a computer model, or a just-so story.  When you actually try to build something you can’t hide from your ignorance. It either works or it doesn’t.”

“The history of aviation is relevant,” says Venter. “Planes would fall out of the sky and there would be intense investigations of why, and then you’d correct it the next time.” He mentioned planes at a recent talk in Seattle. “Imagine designing them if you didn’t know what 20 percent of the parts did,” he told the room. His uncle, a lead designer for the Boeing 757, shouted from the back of the room: “What makes you think we knew?”