Skip to Main Content

The announcement today of a project to synthesize the human genome from off-the-shelf parts is sure to prompt a lot of hand-wringing among scientists and other critics fearful of a future full of lab-grown humans.

But such a science fiction scenario is not what the work is all about, researchers and entrepreneurs involved have told STAT. Synthesizing genomes should help scientists better understand the fundamentals of biology and disease, they said.


Here, STAT looks at the promise (and perils) of building a human genome from the ground up.

Why do scientists want to do this?

“We won’t really understand the genome until we can build it up from scratch,” said Dr. Matthew Porteus, a cancer researcher and gene-editing expert from Stanford University who spoke on a panel at the May 10 meeting at Harvard Medical School.

The genome contains the instructions for life. And while scientists can decipher the complete sequence of DNA that makes up a human genome, trying to build the 3 billion nucleotides letter by letter would be a test of how much we know about it.


“Moving beyond reading DNA to writing DNA is a natural next step,” Francis Collins, director of the National Institutes of Health, told STAT in an email.

Pamela Silver, a Harvard bioengineer who was invited to but did not attend the meeting, said that taking the genome apart (with new gene-editing techniques like CRISPR) and putting it together (through synthesis) are “complementary” methods: “No one is better than the other.”

What can we build so far?

Twist Bioscience, a California-based synthetic DNA company whose CEO presented at last month’s meeting, can currently build made-to-order strands of DNA that are about 1,800 base-pairs long, the length of some individual genes.

Scientists have used these kinds of DNA fragments to stitch together a bacterial genome and part of a yeast chromosome. But human genetic material presents more challenges.

For one thing, a human genome is much, much bigger, consisting of more than 3 billion base pairs. By comparison, the synthetic bacterium created by the J. Craig Venter Institute in 2010 had a genome that was three hundredths of one percent that size.

Then there’s all the layers of epigenetic regulation. The human genome is full of chemical tags that turn genes on and off at different times — something not found in many yeast genomes. Plus, the DNA is wrapped around proteins called histones, which help keep the genetic material stable, and then packaged up to form chromosomes.

Alina Chan, a postdoctoral researcher in Silver’s lab who gave a presentation about her work at last month’s meeting, is in the early stages of trying to put all these elements together to build and test human artificial chromosomes, the first step in making the real thing. But, she said, scientists are “not actually very successful at building these things yet.”

Synthesizing a human genome just “isn’t a technically feasible thing today,” said Rob Carlson, managing director of the biotech investment firm Bioeconomy Capital. As such, the research community has plenty of time “to figure out how to get started, or if we should get started,” he said.

“I think that’s the first thing on the table: Should we do this?”

So … should we do this?

Attendees at the Harvard meeting said this was an active topic of discussion. Coorganizer and Harvard geneticist George Church said that two panels were dedicated to discussing ethical questions, which he told STAT were “some of the high points of the meeting.”

Porteus participated in one of those panels and raised a number of ethical questions, including who would define the goals of this project. In the Human Genome Project, the goal was relatively straightforward: to sequence all 3 billion base pairs.

“It’s a little less clear what the endpoint of this project would be,” Porteus said.

The meeting has come under criticism from individuals who did not attend — one of whom was invited, some of whom were not — for discussing behind closed doors what they argued should be a public matter.

When can we expect artificial human beings?

“Let’s avoid leaping to fictional scenarios, which are neither scientifically feasible nor socially acceptable, and instead contemplate how to catalyze significant advances for understanding how life works and how disease occurs,” Collins told STAT.

Chan said that even if scientists were able to build a human genome from the ground up, it would be a far cry from a real person.

Synthesizing a genome is like writing a story, she said. You put together different elements in a certain order with connections and transitions. Through the writing process, you learn what works to make a good story, and what sounds trite and cheesy.

And by studying great literature, you can learn about the human condition. The story may inform your behavior, but the best-written characters, no matter how realistic, don’t come alive.

“Being able to write a book doesn’t mean the story actually becomes real,” Chan said.

This story was originally published on May 18. It has been updated.