George Church, PhD, of Harvard Medical School

When George Church, PhD, and co-founders established Colossal Biosciences, they envisioned applying their advancements in CRISPR gene editing to the loss of biodiversity, species preservation, and species restoration worldwide.

To expedite the gene editing and genome assembly, Church and colleagues started to build a software platform along with machine learning models.

“We quickly realized that what we had built really to connect Harvard and Colossal and advance the cutting-edge CRISPR gene work being done there, that that platform and machine learning models could apply to the entire synthetic biology market,” Kent Wakeford, who until recently was Colossal’s chief operating officer, told GEN Edge. “Whether that is biomanufacturing medicines, biomaterials, biofuels, recyclable plastics, or crude alternatives, the platform could help advance all of that science and all of that innovation forward.”

“That’s when the light bulb went off for us, and we decided to spin out from Colossal, raise capital, build up a team, and really make this available to the entire synthetic biology ecosystem,” Wakeford recalled.

Wakeford is now co-CEO of that spinout, Form Bio, sharing the helm with Andrew Busey. The two head a company that hopes to replace the conventional, code-heavy software still used by many scientists with a user-friendly platform intended to help them do their work faster and with less effort.

The company projects that its platform is poised to save life sciences customers “millions of dollars” within the first year of being deployed.

“Our platform is generally applicable to anyone within synthetic biology to do their research, to do their analysis, and their validation work,” Claire Aldridge, PhD, Form Bio’s Chief Strategy Officer, told GEN Edge.

The platform is especially applicable, she added, to gene and cell therapy companies advancing into the manufacturing process.

“We can optimize the construct that they’ve created and make it cost a lot less money, have better results, and be safer within the human body. And no one has done this today,” Wakeford added. “One area that we’re so passionate about is candidate validation, so that’s going to be our core sweet spot to go after.”

Form Bio recently launched with an oversubscribed $30 million Series A financing.

Growing market

The global synthetic biology market is expected to enjoy a compound annual growth rate (CAGR) of 27% through 2028, more than tripling to $32.73 billion from $10.11 billion last year, according to Vantage Market Research.

Software platforms are a growing component in the development of synthetic biology drugs and other products, judging from announcements in recent months.

On October 25, synbio drug developer Abalone Bio announced a collaboration of undisclosed value with Danish bioinformatics cloud platform provider PipeBio, which agreed to offer analysis designed to complement Abalone’s Functional Antibody Selection Technology (FAST) antibody discovery platform. Abalone aims to discover treatments for G-protein coupled receptor (GPCR) and other hard-to-drug membrane protein targets, in disease indications ranging from inflammatory diseases, metabolic diseases, and cancer.

And in July, customers of synbio and genomics company Twist Biosciences gained access to DNAnexus’ biomedical data analysis platform for use with Twist’s NGS Methylation Detection System and Unique Molecular Identifier (UMI) Adapter System. The partnership, also of undisclosed value, is designed to help Twist accelerate research and reduce its time to design, validate, and commercialize new next-generation sequencing (NGS)-based diagnostic tests.

DNAnexus’ Chief Revenue Officer, Mark Swendsen, joined Form Bio earlier this month in the same position, with responsibility for all aspects of establishing and scaling the spinout’s sales team and building a global customer footprint. Before DNAnexus, where he led company acquisitions and the sales organization, Swendsen held positions at value stream management platform developer Plutora, eCommerce fraud protection software solution Signifyd, and business continuity/disaster recovery software company Axcient.

Twist’s CEO and co-founder Emily Leproust, PhD, was among guests discussing “The Future of Synthetic Bio” during GEN’s recent virtual conference “The State of Biotech.”

Form Bio is the first of several planned independently-operating technology spinouts from Colossal, which applies CRISPR gene editing toward de-extinction, specifically restoring the woolly mammoth to the Arctic tundra, thus revitalizing the Arctic grasslands and combatting climate change.

At Colossal, Aldridge said, researchers had some of the tools needed to carry out computational biology tasks, from designing guide RNAs to figuring out which base editing system was the right one.

Addressing the data deluge

Form Bio is aiming its platform at researchers still scrambling to deal with the ongoing data deluge.

“Whether it’s genome sequences or microarray analysis or RNAseq,” Aldridge observed, “what we have are these massive amounts of data, and not really systematic and well adopted ways to look at that data, and analyze that data, and understand the conclusions that you need to draw from that data to: 1) understand its path forward into the marketplace, or 2) help understand what your next experiment needs to be.

Scientists at Colossal addressed those challenges by figuring out a solution in the form of an algorithm that may be shared on the open-source software development platform GitHub.

“There wasn’t a platform that allowed for kind of standardization of that optimization of that, as well as capturing all of the info that you would want if you were going to do this over and over, [namely] the citations that you might want to know that this has been done and cited in other work,” added Aldridge, a member of Colossal’s Scientific Advisory Board.

She saw the same challenges that vexed Colossal’s researchers arise in her work as a biotech venture capitalist at Remeditex Ventures, and as a tech transfer specialist holding several positions at the University of Texas (UT) Southwestern Medical Center on and off between 2005 and 2021.

“What we recognized is that those problems are a little bit universal, and the bioinformatics cores are overwhelmed,” Aldridge said. “A lot of times, the rate-limiting step to your science becomes being able to get someone in the bioinformatics core to help you with the analysis of your data.”

The solution, Aldridge and colleagues reasoned, was to develop a platform that allowed researchers to upload data, share it with collaborators, have validated workflows for functions that enhance understanding of data such as RNAseq analysis, and facilitate visualization of that data for purposes that included publications, patents, grant proposals, and conversations with regulators.

Form Bio’s platform integrates core components of data management, workflows, results visualization, and collaboration into a single solution. The platform also includes a catalog of verified workflows extensive enough to cover a variety of scientific uses, from ancient DNA analysis to transcriptomics to adeno-associated virus (AAV) gene therapy.

The platform is also intended to serve as a foundation for advanced, AI-based applications tailored to the needs of specific industries and academic fields.

“If we were able to build that for Colossal, we could build that and then allow the entire biotech industry to leverage that, and basically let the scientists spend more time doing science and looking at the data to inform their next experiment or the next step in the path,” Aldridge said. “We could let the bioinformatics do the things that are really more bespoke instead of those analyses that they end up doing over and over, that can be automated with these workflows.”

Questions and menus

For RNAseq analysis, as an example, the platform walks users through with a series of questions and drop-down menus: What genome do you want to be your reference genome? What analysis do you want to run? And how do you want to see that displayed?

“It really walks people through the process and allows them to iterate within that process to say, Well, I tried it against this reference genome. Do you know what? It might be interesting to see it against a different reference genome, and see how that might change the interpretation of that data. That’s a big part of it,” Aldridge said.

She said she and Form Bio colleagues are especially excited about starting to apply machine learning and AI models to understand manufacturability.

“The first indication that we want to go after is, how we design these molecular medicines for ease of manufacturing? When you think about what goes on now in cell and gene therapy, and other things like that, mRNA vaccines, often that first construct that the researcher, whether they’re at a company or an academic environment, does [is], if they see a signal with that, that’s what goes into development.

“There’s never that step to say, How do we look at this and make sure that there’s not weird hairpins in the DNA sequence that’s going to cause truncation when we put it in the big bioreactor? How do we know that there’s not some weird other odd sequence anomalies in there that are going to lead to truncated proteins or CPG Islands that the innate immune system would recognize? And so how do we do that in a way that allows us to get the most drug and the safest drug out of that manufacturing platform?

By reducing hairpin formation, Aldridge said, users of the platform look to generate more viral capsids filled with full-length constructs, not truncated or empty ones, “because that leads to more doses, and that leads to a lower cost of goods and that improves the safety profile as well.”

At deadline, the only user of the platform Form Bio has disclosed is Colossal, though the spinout says it is in conversations with other interested potential customers.

Church, a genomics and synthetic biology pioneer, and co-founders conceived of Colossal in 2006, but the company wasn’t formally launched until September 2021, after it completed a $15 million seed financing round. In March, Colossal secured an additional $60 million through a Series A financing.

Before Colossal raised the $75 million, PayPal co-founder and billionaire entrepreneur Peter Thiel invested roughly $10,000 a year for 10 years in the company, Church said last year, speaking on GEN Edge’s “Close to the Edge” video interview series: “That’s not the sort of [financing] you can do this kind of challenging project on, but it did keep us thinking about it.”

“It’s a fun project, because you really have to think in a very interdisciplinary way and have to be thoughtful about all the economic and social consequences,” added Church, director of Synthetic Biology at the Wyss Institute, a Professor of Genetics at Harvard Medical School, and Professor of Health Sciences and Technology at Harvard and MIT.

Church serves on Form Bio’s 14-person Advisory Board, whose members include Nobel laureate Carolyn Bertozzi, PhD, a co-recipient of this year’s Nobel Prize in Chemistry for the development of click chemistry and bioorthogonal chemistry.

Mammoth undertaking

Through a sponsored research agreement, Colossal supports research in Church’s lab to create an elephant-mammoth hybrid that is genetically engineered with traits designed to help it survive in the Arctic.

Computer-aided design, fabrication, testing analyses and machine learning are key to the future of bioengineering in general and specifically restoration of endangered and extinct genetic diversity for keystone species in vital ecosystems. Form Bio is the software critical to pave the way,” Church said in a statement. “As scientist-engineers, we need these pipelines and look forward to faster breakthroughs in scientific discoveries and applications, now that software has caught up with science.”

Where does Colossal’s work end, and Form Bio’s begin?

“Colossal is doing cutting-edge biology. They are on the cutting edge of CRISPR gene editing to solve big issues facing the planet,” Wakeford said. “Where Form helps is in the computational and algorithmic work that helps them predict what they’re going to do, and give them more of a roadmap once they’ve done it, help validate it, and really shrink the time that it takes to get from Point A to Point B in creating the mammoth or any other species.”

The sizes of the workforces are comparable, for now. Colossal has 45 PhDs based in Boston and Dallas, where Form Bio is based. Form Bio is starting out with about 40, with plans to nearly double: “We anticipate being about 70 employees by the end of next year,” Wakeford said.

Most of those new employees will be based in Austin, TX.

“A big focus of our hiring will be around machine learning and product development, specifically around these really innovative cutting-edge tools that we’re creating,” Wakeford said.

JAZZ Venture Partners led Form Bio’s $30 million Series A round, with participation from Colossal’s lead investor Thomas Tull, founder and former chairman and CEO of Legendary Entertainment; and another investor in Colossal, Builders VC.

Form Bio needed just three months to close on its financing. Talks started a week after the company spun out of Colossal on July 5.

“There were a number of investors in Colossal who have been watching the progress on our software from the early days. And Jazz and Builders VC have been following our progress and believe that what is being built at Form can have a real meaningful impact on advancing science,” Wakeford recalled. “As soon as we spun out, we immediately had conversations with them. And based on those conversations, they decided to lead the round, which was fantastic. They’ve been just exceptional partners for us.

“While there is a lot happening in the overall market and economy, I think, because the investors who led this have been watching the progress and the team for so long, they were able to have a very informed thoughtful decision, and be able to make that relatively quickly,” Wakeford added.

Previous articleToxoplasma Parasite Persuades Immune Cell to Change Behavior, Aiding Dissemination
Next articleComputational Approach Gives Detailed Map of Human Tissues