Last July, scientists created the first “synthetic cell,” an organism that’s controlled by a chemically synthesized genome edited on a computer and stitched together in the lab. One year later, biologists at the Fifth Annual Synthetic Biology conference at Stanford University are still struggling to take the next step in the field. Holding them back are the vagaries of biology itself, and the expense and time needed to get from idea to engineered organism.

While the creation of the synthetic cell, at the J. Craig Venter Institute, hints at a future in which synthetic biologists can redesign living cells to perform whatever tasks they dream up, that goal is still distant. Most research has focused on coaxing microbes to perform tasks that are similar to what they already do, such as transforming sugar into fuels using processes and materials that resemble the ones they use in nature.

Synthetic biology strives to make molecular biology more like engineering—with predictable materials and parts that can be put together in predictable ways. As the synthetic cell demonstrates, scientists now have the tools to edit an existing genetic sequence on a computer, use DNA-synthesizing machines to create it in fragments, and stitch these together in the lab. (This route is just one of many that synthetic biologists are taking.) But it’s still difficult to predict what cells will do after they’re altered. Researchers are often stymied by cells’ natural drive to grow and live as they please, which in many cases must be overcome to get them to do something useful in an efficient manner.

One of the biggest hurdles lies in the creation and assembly of starting materials: modular bits of DNA that code for a particular function and are synthesized in the lab. Creating this DNA is time-consuming and expensive. Like any commercial product, it must be designed, built, and tested. Even making relatively small changes can take a lot of work, a long time, and a lot of money.

“Some sequences take two months to synthesize,” while others can’t be made at all, for reasons that are not yet understood, said Reshma Shetty, cofounder of Ginkgo Bioworks, a startup that assembles DNA parts. Shetty said the company uses software-based automation to design building blocks and other parts, and to control liquid-handling robots that mix them together from pieces of DNA ordered from other companies that specialize in DNA synthesis. It’s this last step that’s currently a major bottleneck. The company has been tracking how long the sequences take to make and which sources do it fastest.

The expense and time involved in creating new organisms limit creativity, said Pamela Silver, a professor of systems biology at Harvard University. Every time synthetic biologists try out a new design, they have to pay to get the DNA synthesized, wait for it to come back, get it into cells, and test it. All this, says Silver, means synthetic biologists are understandably reluctant to fail and learn from their failures.

“I still believe in the dream that some of you will eventually be able to sit at a computer, design an experiment, and get the DNA the next day,” she told the crowd. For synthetic biology to deliver on its promise, DNA synthesis needs to be “cheap, fast, predictable, and accurate—and open to all,” including researchers whose labs don’t have a lot of equipment or funding.

Fortunately, the cost of DNA synthesis technology, much like that of DNA sequencing technology, is dropping rapidly. George Church, director of the Center for Computational Genomics at Harvard, noted in his talk that the costs of both DNA synthesis and sequencing technologies have been decreasing at an astonishing rate—lately by a factor of 10 each year.