Since the 1980s, scientists have relied on chemical synthesis to build custom DNA molecules base by base. It’s been reliable for synthesizing short oligonucleotides less than 200 base pairs in length. However, it uses a large amount of organic solvents, phosphoramidites aren’t stable on the bench, and it’s not suited for synthesizing long pieces of DNA. With an assembly efficiency of ~99%, phosphoramidite synthesis methods for oligonucleotides larger than 200 base pairs means only 13% of the oligonucleotides produced are correct. 

The 2000s saw the rise of enzymatic synthesis and Gibson assembly-based methods to piece together longer stretches of DNA. In 2021, a synthetic DNA molecule longer than 10 kb was created for the first time. The company behind this feat…Ribbon Biolabs.

In this Q&A, I spoke with Jodi Barrientos, CEO and CBO of Ribbon Biolabs, to learn more about long DNA synthesis and how Ribbon’s technology is able to create synthetic DNA up to 20 kb in length.

Additionally, this is a topic of conversation that will be discussed in detail at SynbioBeta 2024: The Global Synthetic Biology Conference, May 6 – 9 in San Jose, California as part of the Reading, Writing, and Editing DNA track.

‍

What is the motivation to synthesize long DNA? How did Ribbon get started?

Jodi Barrientos: Ribbon’s history is one of necessity. Right now, you can capture about 39% of the genes with the technologies on the market. It's not enough because we know biologically, there are a lot of really important genes that are both complex and long. With Ribbon’s technology, we can now capture about 61% of the genes available. We're continuing to push those limits, and that's just on the length capability. That's really what drove the necessity of creating Ribbon, increasing the available number of genes we can synthetically manufacture for ease of interrogation in synthetic applications. 

‍

What sizes of DNA is Ribbon manufacturing? How long can you go?

JB: Right now, we are able to manufacture up to 20 kb without issues. That's where customers have asked us to go. We've done a pretty exhaustive survey of the market, and we found that right now, the need is really between 10 and 15 kb. That doesn't mean that's where we think our technology can stop. 

As people start thinking about these longer genes, we think that we're going to see expansion beyond that. Length is one of many areas of molecule complexity that customers can't be served today. Because of our technology, not only can we get to the length, but we can also get to the corner cases of complexity like GC content, homopolymers, and repeats. 

‍

Can you tell me more about how Ribbon’s assembly algorithm works?

JB: We take into consideration the fact that molecules are unique and molecules are complex. Our machine learning algorithm reads through the sequence of interest that the customer provides to us. Based on our core technology, a set of very small oligos, and a series of enzymes that are proprietary to Ribbon, the algorithm determines how to best assemble the specific sequence. 

This gives us the ability to get around complexities where some other technologies can't go, and it allows us to be exceptionally accurate. This accuracy becomes extremely important as cancer vaccines and cell and gene therapies come into the market. Ribbon's approach enables the identification of errors in quantities as low as 0.33%, whereas competitors can only detect and report to 10%.  This means our customers can be very confident about the molecules in their tubes.

‍

What is the assembly process like?

JB: We start with short building blocks because starting with short, accurate molecules allows you to build long, accurate molecules. We have nearly 65,000 short oligos in-house that we can pull from those, and we start to assemble the molecule right off the line. This is something that helps us fuel the ability to produce molecules in real time.

From there, it’s an enzymatic assembly. We use proprietary enzymes to bring assembled fragments together into full-length molecules.

‍

Is this process completely automated? 

JB: You can imagine that trying to do this by hand would be chaos. And so we automate the full end-to-end process, and it's driving forward on a production scale.

‍

Using methods like chemical synthesis, you can add fluorophores, other tags, and various modifications. Are there any challenges to doing similar modifications with long DNA synthesis?

JB: We have not tested it, but there's nothing we feel would prohibit it from a technological standpoint. Once we see that there's a market demand there, it's something that could end up on our product roadmap. We think that the technology could do that. It's just a matter of we haven't gotten enough demand to drive it forward.

‍

Can you speak on some of the applications of long synthetic DNA? Where is it most needed?

JB: One of the most attractive things about long, complex DNA synthesis is that there are just so many directions you can go with the technology. Everyone who's using synthetic biology is going to need longer and better molecules, and that's regardless of the application.

The early organic demand that we've seen is coming from biopharma. We know that this is a huge market because the direction that biopharma is going is towards synthetic biology applications. This is where we can change the dynamic for biopharma customers in areas like biotherapeutics, cell and gene therapy, mRNA production, and viral vaccine production.