The Next Frontier in Gene Synthesis: Conquering Complex DNA

Scientists have developed a new technology that will significantly expand the design space that can be tested with synthetic DNA
Emerging Technologies
Dan Lin-Arlow, PhD
May 1, 2024

For all the promise of synthetic biology, scientists are surprisingly limited in their ability to explore the full genetic design space of possibilities—whether that’s for developing a therapeutic candidate, modifying biological fibers to make a new material or engineering yeast to manufacture important molecules via fermentation. For any of these bread-and-butter synthetic biology pursuits, having the ability to test as many types of candidate DNA sequences as possible is key to finding the most effective one.

Today, the single biggest constraint in synthetic biology comes from the starting point for virtually all experiments: obtaining synthetic DNA. Fortunately, scientists at Ansa Biotechnologies are developing a novel technology designed to address the biggest pain points facing DNA synthesis.

The promises of rapid and reliable delivery by the dominant DNA synthesis vendors give the impression that scientists can design any DNA sequence and have it on their bench in a few days. In reality, there is a litany of limitations around which sequences these companies can produce. Vendors call the sequences that are difficult for them to make “complex.”

So what, exactly, is complex DNA? While there is no universally accepted definition or technical guideline, complex DNA is generally understood to mean sequences that contain regions of high GC content, homopolymers, hairpins, and repetitive elements, among other features. These sequences will have a high chance of failing at some step in the vendors’ complicated manufacturing workflows.

The largest DNA synthesis vendors reject most complex DNA sequences. Sometimes, a vendor will attempt to build the DNA but later notify customers that the sequence failed in production. Researchers will often be directed to specialty shops that are more adept at making complex DNA constructs. While it is helpful that these shops exist, using them typically entails very long timelines and price points that are prohibitive for most scientists, and many important sequences remain out of reach of existing artisanal methods.

The more common response to the challenge of ordering complex DNA is to abandon the desired sequence. For coding sequence, researchers might use codon optimization to tweak their sequence of interest, adjusting it into something with lower complexity that can be synthesized. At best, the output of this process is a proxy for what the scientist originally wanted to study; the optimized sequences may differ functionally from the initial designs even though they code for the same amino acid sequence. When codon optimization fails to yield an easily synthesized alternative to the desired sequence, scientists generally must replace it with another sequence that might be less promising or less suitable for the research goal.

In this way, and in laboratories worldwide, the design space is pared down substantially from the territory scientists would like to explore to the narrow confines of what can be explored right now. The sequences tested today represent a fraction of the sequences that should be tested for optimal results.

To be sure, there may always be some sequences that cannot be synthesized—or, if they can be synthesized, they couldn’t be delivered to scientists because the sequences cannot be amplified or cloned. For example, sequences containing a stretch of 500 adenine bases (that could serve as a template for mRNA production) are quite difficult to amplify and to propagate stably in cells. But many, or possibly most, of the complex sequences that stymie conventional vendors should be possible to synthesize and deliver to help scientists accelerate their research and achieve better results.


A new and unconventional approach to DNA synthesis could solve this problem. In Emeryville, Calif., scientists at my company, Ansa Biotechnologies, are building on an idea born in Jay Keasling’s lab at the University of California, Berkeley. To overcome the challenge of synthesizing complex DNA, my co-founder Sebastian Palluk and I believed that we had to look beyond the traditional synthesis methods based on phosphoramidite chemistry. That process damages DNA as it is built, inherently limiting the final product’s quality and, crucially, its length, forcing vendors to stitch together dozens of short molecules into the full-length sequences that customers request. The problem is the stitching process cannot reliably assemble complex sequences.

Instead, we developed an enzymatic DNA synthesis method to avoid these inherent restrictions. Our technique harnesses the naturally efficient DNA-building capability of the enzyme terminal deoxynucleotidyl transferase (TdT) to synthesize customer sequences directly—no stitching required. The method is based on polymerase-nucleotide conjugates, essentially TdT molecules that have been pre-loaded with a single deoxyribonucleoside triphosphate (dNTP) molecule that they can use to rapidly extend a DNA molecule by exactly one base at a time. Importantly, our method operates under mild conditions and does not damage the growing strand during synthesis or place any meaningful limitations on its length or sequence. This allows us to generate long, accurate DNA strands for building DNA constructs and provides access to far more complex sequences than can be readily stitched together using conventional methods—all while getting this kind of DNA into scientists’ hands faster than ever.

There are many types of sequences that are important for the next generation of life science research that are difficult for scientists to obtain, including promoters for engineering gene expression, CRISPR arrays for genome editing, and ITRs for gene therapy vector development. Enabling the synthesis of these complex DNA sequences will provide the foundation for future advances in all of the fields impacted by synthetic biology, including biologic drug development, biomaterials, ag tech, and fundamental life sciences research.

Ansa’s technology is now available through an early access program. To learn more, visit ansabio.com/early-access.

Dan Lin-Arlow is co-founder and chief scientific officer of Ansa Biotechnologies.

Related Articles

No items found.