Asimov – Building Tools to Program Living Cells
Table of contents
Blockbusters – drugs with annual sales of at least $1 billion – are to pharma companies what unicorns are to venture capitalists (VCs). They’re what you ultimately want to find. Proteins are the basis for many blockbuster therapeutics. For example, Insulin, the protein-based hormone used to treat diabetes, had a market value of $21.2 billion in 2018. Proteins as medicines are a massive opportunity as many diseases and biological processes in the human body are governed by the presence (or absence) of these molecules. However, right now, most pharmaceutical companies have to go through the laborious process of screening potential protein candidates for their possible therapeutic value and there are few ways to intelligently construct proteins for their targeted effects.
Basically, we’re stuck with what we already know from research studies of proteins that already exist in the human body. Proteins that we recognize have an effect, like hormones, enzymes, antibodies, and peptides, are what biomedical researchers have squared away in their toolbox of knowledge. The best we can do is make some minor genetic tweaks to these pre-existing proteins, grow them in the laboratory, and be sure they can pass through the Food and Drug Administration (FDA) approval process without too many side effects. Manufacturing proteins is already costly, so treading too far from the well-trodden path can lead to higher risk of failures and huge losses.
But what if there was a way to engineer completely new therapeutic proteins from scratch, one that wasn’t already designed by Mother Nature and the laws of human evolution? The below chart gives you an idea of just how many possible proteins are lying in wait to be discovered.
It’s something we covered in last summer’s piece on Designing Proteins to be Molecular Machines. Today, we’re going to introduce you to a life sciences company that’s developing a full-stack solution for programming cells to do useful things for mankind.
A Full-Stack Genetic Circuitry Startup
Bahstun-based Asimov is a bioengineering startup co-founded in 2017 by Chris Voigt alongside two of his students and Douglas Densmore, a computer engineering professor at Boston University. Chris Voigt is a big-time MIT professor and pioneer of synthetic biology who also co-founded the MIT-Broad Foundry for Synthetic Biology. Reading his resume makes you realize how little you’ve accomplished so far during your short time here on earth. As a graduate student at Caltech, he trained under the 2018 Nobel Prize winner, Frances Arnold, who invented the processes used for directing the evolution of biological systems to produce next-generation enzymes, therapeutics, and biofuel. Asimov has raised a total of $30 million in disclosed funding following a $25 million Series A last summer which was led by notable VC firm Andreessen Horowitz.
The company is building a full-stack genetic circuit platform. Basically, all cells use strands of deoxyribonucleic acid (DNA) as the master blueprint for life. What a cell does and what it produces is tied to the four-letter (C, G, A, and T) code strung together in the right sequence to produce whatever protein you want. Using a combination of machine learning, genetic engineering, and cellular manufacturing, Asimov is creating a way for companies and organizations to construct proteins on demand by digitally reprogramming cells in the same way that electrical engineers can build electronic devices using computer-aided design (CAD) software. Asimov was selected by DARPA in 2019, the same defense agency that developed GPS, the autonomous vehicle, and, of course, the internet, to design a physics-based artificial intelligence engine for engineering living cells.
Using Cellular Bioengineering to Build Intelligent Therapeutics
Genetic engineering is like the female mind – we’ll never truly understand it because it’s extremely complicated sophisticated. While inserting DNA into a cell is already a challenge, making sure that DNA does what you want is the other half of the problem. Even if a cell produces the right protein coded in the DNA, other elements can run interference and the target protein may come out damaged, misfolded, or in low amounts. Sometimes the protein can even kill the cells. And mammalian cells, the ones that make up you, me, and Fido, are even more prone to errors compared to simple creatures like bacteria and yeast because these cells contain way more information, processes, and molecules.
But that’s where the world’s next blockbuster protein therapeutics need to be produced, because mammalian cells contain the correct machinery to properly fold, tag, and package these therapeutics in a way that’ll make them highly functional in the human body. Bacteria and yeast can only do so much to keep up with their simple biological structure. Imagine trusting Domino’s to design, bake, and deliver a world-class pie that rivaled a recipe from award-winning Tony’s Pizza Napoletana, and you get the picture.
Enter Asimov. The company’s platform designs genetic circuits, which are a series of genetic codes that work together to ensure a protein is manufactured in the right way. Much in the same way as electrical circuits are used to connect different elements together to create a device like an alarm, a television, or an iPhone, a genetic circuit can be used to string together biological connections in a cell to program it to produce a protein. A simple circuit could contain a handful of genes, which is already challenging for synthetic biologists to put together. But proteins that require several genes jammed together can get out of hand.
The founders of Asimov published a paper in Science back in 2016 titled “Genetic circuit design automation”, where they outlined their forerunner to the Asimov platform called Cello, short for cellular logic. With co-founder Alec Nielsen as the lead author, the paper was cited over 600 times, which is similar to getting likes from the scientific community, except “posts” require years of brain-wracking work and most research papers are lucky to get even ten citations. Not quite the same as being an Instagram influencer, someone who subtracts value from society.
Using the genetic circuitry of the common E. coli bacteria as a prototype, the founders outlined their philosophy behind Cello and Asimov. The current paradigm is to tinker with DNA by manual trial-and-error at the level of a single genetic “byte”. The Cello platform turns the idea on its head and uses large sections of DNA to create interchangeable parts like strings of code that can be melded together automatically using machine learning, much like Lego pieces. The platform even predicts how well those circuits are going to work inside a cell. Using the Cello platform, the research team rapidly designed genetic circuits that doubled the record for the largest circuit constructed by hand, which was done back in 2009.
Use Cases for a Genetic Circuit Platform
The most obvious use case for the Asimov genetic circuit platform is constructing therapeutic proteins with lower lead times and higher efficiencies. The platform can be used to rapidly construct a massive library of proteins with slight modifications to their molecular structure to identify if new variants on pre-existing protein therapeutics or candidates have better biological properties than their native sequence. These would include antibodies that are now used in cancer immunotherapy, where the antibody could be optimally designed to suit a specific population’s cancer type. The platform allows drug designers to iterate quickly and evolve their protein products to achieve a desired target effect, stability, or solubility.
The technology can also be used to optimize for proteins used in other sectors. Protein-based mammalian hormones are a key ingredient in cell-based cultures and are some of the more costly components. A genetic circuit platform could be used to precisely design, test, and manufacture hormones for growing meat in the laboratory with improved efficiency and growth. The platform could also be used to engineer new types of mammalian cells with taste properties that have never been experienced before in nature. The growing intersections between the cellular agriculture industry and advancements in synthetic biology could evolve into a powerful synergy. Get ready to eat tiger meat that tastes like cheddar cheese.
More precise control over the genetic equipment of a cell would provide new opportunities to design biological systems, tissues, and even organs. Right now, we have limited control over the types of cells we can generate using stem cells. Asimov’s platform could allow us to grow new types of tissues and organs we never had before, with completely different properties from what we evolved naturally as humans. Imagine constructing an organ that’s fully integrated into an electronic device because the genetic engineers were able to integrate the expression of metalloproteins and scaffolds in the cells with the metallic circuitry of the device. Bioengineers could pull out the gene sequences for the quantum mechanical proteins used by birds to navigate during migration and embed them into a human to create a new type of human sense. Victor Frankenstein would be proud.
Conclusion
While Asimov is still a small company, what the team and other groups in this space are proposing is nothing short of revolutionary. Essentially, they are building the foundations of a biosynthetic computer and ripping out the genetic logic found inside cells to create their own programming language that speaks 100 percent bio. If the discovery of DNA was like the discovery of the semiconductor, and the Human Genome Project, which set out to sequence the entire human genome, was like the discovery of programming, Asimov is constructing Microsoft Windows for the Human Genome Project. Except maybe Asimov’s platform won’t give their users the genetic equivalent of the blue screen of death.
Share
