The progamming of living things: What if you could put a brain inside a cell?  : Biofuels Digest

0
204


Illustration of the molecular milieu inside a white blood cell. Cells are complex biochemical entities capable of sophisticated computation. (David Goodsell)

Think of it. There’s the brain — with its trillion synapses, sensing, diagnosing, prescribing, responding, targeting. What if a human cell could be programmed to detect a disease, manufacture a drug and deliver it. As in no doctor, no skyrocketing drug cost, no delay, no hassle, no kidding.

It would revolutionize medicine. But also, agriculture — as in plant cells that sense and respond to conditions like drought, predators, disease or even the changes that nightfall brings. Not to mention industrial biotechnology — as in cells that sensed and responded to changing conditions in fermentation vats.

The good news is that you won’t have to wait too long. The fundamental breakthroughs have happened.

Now, we have a programming language for bacteria (or any micro-organism, eventually), that builds bio-circuits that add intelligence to cells. The first appearance of an industrially-relevant language appeared about 22 months ago in this issue of Science, in which they described the ability to to build circuits that can detect up to three inputs and respond in different ways.

Using this language, anyone can write a program for the function they want, such as detecting and responding to certain environmental conditions. They can then generate a DNA sequence that will achieve it.

“It is literally a programming language for bacteria,” says Christopher Voigt, an MIT professor of biological engineering. “You use a text-based language, just like you’re programming a computer. Then you take that text and you compile it and it turns it into a DNA sequence that you put into the cell, and the circuit runs inside the cell.”

Where we were before this

“What we’ve had in biotechnology is the constant overproduction of a few proteins,” Asimov CEO Alec Nielsen told The Digest. “And proteins are incredible, these anatomically precise nanostructures. But cells are a higher order altogether. Multiple genes, multiple proteins, genetic circuits engineered to have some function. Sensing the environment, responding to those signals, controlling gene expression. Think of the cell as a medicine manufacturer and delivery system, if you can program it.

“The brain is a device, a much higher order, trillion synapse computational device. But imagine giving a cell a brain, even if for a much lower order of operation. Then, we can turn our eye to therapeutically and industrially relevant targets, even more so at Asimov than the original team at MIT which is focused on academic concerns.

The lac operon genetic circuit. In response to glucose and lactose availability, E. coli regulates the expression of genes involved in lactose metabolism. (Wikimedia Commons)

And now, a company called Asimov has spun out of MIT aimed at detecting and responding to  customer needs in applying this capability to real-world opportunities — instead of the academic focus of the work which continue at MIT and elsewhere. Asimov just picked up a $4.7 million seed round investment from Marc Andreessen’s’ venture firm. You might remember that Marc was a hot-shot spin-out CEO himself a generation ago, developing something he called a web browser, which, um, did pretty well.

Inspired by the trajectory of electronic design automation — they’re making the engineering of biology follow the same workflow of engineering a computer chip. With Asimov, a biological circuit design starts in the very same way that a computer chip design would start: by programming it in Verilog, the language used to design electronic circuits for decades.

The Third Wave Begins

The first wave in biocircuits was the discovery that you could make them at all, synthetically that is — a discovery in 1961 that landed the 1965 Nobel Prize in Medicine.

The second wave landed on the shore about 18 years ago, following a couple of advances which ultimately found expression in two issues of Nature in 2000 — advances that made it possible for biological engineers to

in Nature that made it possible to design many genetic parts, such as sensors, memory switches, and biological clocks, that can be combined to modify existing cell functions and add new ones.

But it was slooooow. Expennnnsive. And you needed a PhD to do it. Like the days of John van Neumann and Alan Turing building the first digital computers.

These days in digital programming, we have toolkits.

As Andreessen Horowitz general partner Vijay Pande explains, “One does not design every transistor in a modern microprocessor by hand, but instead designs it in modular parts (e.g. circuits to do memory, arithmetic, logic, control, etc.) that are then combined.”

Now, we have a programming language and, via an interface like we see below in Cello, we have the ability to rapidly engineer biocircuits.

Overview of the original Cello platform [7]. A Verilog specification is automatically compiled to a DNA sequence that encodes a genetic circuit.

The simulator

Pande pointed to simulation as a breakthrough aspect of Asimov’s technology,

“The next step — in both the electronic and biological context — is to predict the outcomes of circuits, because making new chips as well as new cells is expensive at the prototype stage. EDA tools include powerful simulators of circuits, so engineers can debug them virtually, resulting in a low-cost, high-turnaround process. Asimov’s custom tools include a powerful simulator that can predict whether or not a biological circuit will work with up to 90% accuracy.”

Modular circuit components

Pande added:

“Not only do such biological circuit design automation tools give bioengineers the ability to debug biological circuits much like we debug software — with complete detail of what the simulated circuit is doing — but Asimov engineers have also developed modular biological circuit components that don’t have adverse reactions to other parts of the cell. “

The result? Now, Voigt explains, “You could be completely naive as to how any of it works. That’s what’s really different about this,” Voigt says. “You could be a student in high school and go onto the Web-based server and type out the program you want, and it spits back the DNA sequence.”

The first targets: think therapeutics

We’re in early days in terms of sophisticated applications that compare to, say, the complexity and power of microprocessors. As the Asimov team observed in a recent blog post:

“We are able to build genetic circuits with up to 10 interacting genes. That’s state-of-the-art for an engineered cell behavior in 2017, but it still pales in comparison to nature. For reference, the E. coli genome employs roughly 300 genes called transcription factors to control metabolism, survival, and replication. Human cells have about an order of magnitude more.

Pande pointed to therapeutics as an obvious target for early-stage applications. “7 of the top 10 drugs today are biologics, i.e., proteins that have therapeutic properties,” he noted. “ These proteins are manufactured in cells at the cost of billions of dollars. Asimov’s technology could drive a dramatic reduction in cost to patients — enabling these drugs to be in the hands of more and more people.

But it goes deeper, farther. Pande speculated:

“Looking even further out, a bolder application for designing biological circuits is one where new cellular therapies could sense disease in the body, perform logic, and drive a precise curative response — like therapeutic, microscopic “bio-robots”.

The business of Asimov

“We will build circuits for our customers and every single one will be a better way for them than constant overproduction.  The natural genetic circuits were hacked together by evolution. When we design, it looks very different. It’s very modular, through our platform Cello, we’ve taken the manual element out of circuit design. The old days of throwing things against the wall and seeing what sticks has ended. It’s not how we design bridges, or computers, either, so its time that we designed circuits this way and use, for example, machine learning instead of brute forcing all the time.

Depending on the complexity we can be 50-90% accurate in terms of a circuit that meets a proposed need, and we are orders of magnitude faster. Even though we still do experimental validation before we ship a product.

The longer term

What about multi-cellular systems? After all, something as profound yet foundational as nitrogen fixation is done by a symbiotic community of organisms using as array of cells.

“We’re already engineering multi-cellular systems,” said Nielsen, and targets such as nitrogen fixation are something we definitely have our eye on. It’s trickier. But also, think about applications for the microbiome.”

The Bottom Line

For some time this year, we’ve been on a campaign to move beyond outdated terms such as “synthetic biology” and instead for people to look towards “digital biology” where biological science ultimately is understood as a profound, self-sustaining, self-evolving, complex, mobile and three-dimension version of information science.

Seen in this light, the programming of biocircuits is an incredibly important branch — but it’s relevance to everyday life and everyday products is just arriving. Once, artificial intelligence was something explored only in advanced science projects. Today, AI is becoming an important tool of everyday applied science.

Asimov is proving that the same applies to biocircuits. They’re going to change, well, you. And they’re going to change therapeutics and industrials very quickly.

At some stage, we begin to wonder what will happen when the back end of custom cellular manufacturing and the front end of metabolic pathway design both become fully integrated and subjected to machine learning. We all know that robots are important and interesting, but what about bio-robotics? That’s an order of magnitude more interesting and that much closer to — I was going to say a replacement for evolution — but rather, a parallel track to evolution that moves much faster and with far more purpose.

Who’s going to look after all this to ensure that the targets are commercial, therapeutic, society-building and beneficial or benign to our civilization? As we’ve seen in software and internet, evil-doers abound.



Original Source