Making Computers Using What?!
How we can make computers out of DNA, enzymes, and even brains!
What if I told you that your body is made up of trillions of tiny computers, each executing countless instructions to determine the shape of your eyes, your bone structure, and even the number of fingers you have? That’s bonkers, right? According to biological computing and its subfield biological computation, this is the case 🤯.
Imagine if we could read, debug, and edit our body’s code. Or, imagine a world where we could harness the natural computational capability of our bodies and cells to augment our traditional computers. Biological computing tries to do exactly this, both by storing information in DNA and by using cells to do calculations. I’ve already written an article covering the basics of biological computing, so for this article, I’ll focus on biological computation, which I find a lot cooler too 😎.
What is Biological Computation?
Biological computation is one of those beautiful words that means exactly what you’d think it means: doing things a computer might do using biology. There are a couple of ways you can do biological computation. You can use DNA inside a cell the way cells already do to respond to their environment, you can use proteins or other small molecules (including DNA) inside or outside of a cell, or you can use cells like neurons to learn and respond to their environment. The cool thing is, there are early examples of all of these approaches and some combined approaches today all over the internet. One example of this is Cello, a free tool to build genetic circuits that you can run in your browser.
How to turn your DNA into a computer
When you think about it, your cells are running programs all the time. Take the pancreas, for example. The pancreas’ main job essentially amounts to “if glucose is above a certain level, release insulin”. Your cells use special proteins called regulatory proteins and special sections of DNA to regulate what genes are turned on and off in a way that is extremely similar to computer code.
One example of this is the role the Bmal1 gene plays in our circadian rhythm. Bmal1 forms a genetic oscillator with a couple of other genes to control our circadian rhythm. To simplify things, I’ll just call the genes Gene 1 and Gene 2. In reality, it’s more complicated than this, but this simplification is good enough. Gene 1 (Bmal1 plus some other stuff) increases the amount of Gene 2 being made by the cell. Once there’s enough of Gene 2, it goes to the nucleus and stops the cell from making Gene 1. This process looks a little like this:
This cycle takes about 24 hours, so the cell can tell what part of the day it is by looking at the level of these proteins. This entire cycle can be summed up as “If Gene 1 produce Gene 2, if not Gene 2 produce Gene 1”. But, could we design systems like this? Well, it turns out we can! I mentioned cello earlier as a tool to design genetic circuits, and that’s what this is. Cello uses a programming language for designing computer chips called Verilog to let you design genetic circuits and run them in many living organisms. Verilog is kinda hard to learn, though, so I wouldn’t be surprised if cello alternatives pop up at some point. This area of biological computing has incredible potential and is just starting to get big, so I can’t wait to see what people do with this next, especially if they combine it with other parts of biological computation.
Computation With Enzymes and Other Small Things
It turns out that genes aren’t the only small thing in our cells that we can use to do cool stuff. The enzymes that our genes produce can also respond to their environment in a way that’s similar to a computer. There are several enzymes that produce something that goes on to affect how the enzyme works. Think about it, if Substance A gets turned into Substance B and C and there’s a lot of Substance B in the cell, the cell should make more Substance C instead of Substance B. While this method is faster since the cell doesn’t have to make entirely new proteins, it’s harder since you have to design new proteins or find some that will do the job.
This is where DNA re-enters the picture. In 1994, Leonard Adleman solved the travelling salesman problem using DNA. The problem is simple; a travelling salesman must take the shortest path between N towns, visiting each town exactly once. The travelling salesman problem takes exponentially more time to solve the more towns there are, so it’s a good test problem for a new type of computing. Since complementary pieces of DNA stick together, Adleman made bits of complementary DNA that represented towns and roads connecting those towns. He mixed all these different bits of DNA, and they formed a bunch of random sequences of towns connected by roads.
Since DNA is small, you can represent almost every possible solution in a test tube. If you can find the best solution in all this mess, you’ve solved the problem! This isn’t feasible for simple problems like the travelling salesman problem with a couple of cities, but for a complicated problem, you could evaluate many possible solutions at the same time, allowing you to solve problems that would take a supercomputer years to solve in just a couple of weeks. The drawback to all of the approaches discussed thus far is that they have to be built for one specific problem, so they aren’t reprogrammable like conventional computers. Neurons, on the other hand, can learn and adapt to their environment in a way conventional computers have difficulty with.
Neuron-Based Computation
Your brain is the most complicated object in the known universe. The computer I’m writing this on has a maximum theoretical processing power of 469.2 gigaflops. The human brain can do100 teraflops! Everything being done with AI today is running on silicon trying to emulate our brain, but why not just use actual neurons?
Well, it turns out we can 🤯. In 2004, a scientist at the University of Florida grew a bunch of rat neurons in a dish and taught them to fly a plane in a simulation. I mean, the obvious application for this is flying small drones or other plane-like objects, but you could theoretically do so much more.
So what’s the delay? Why don’t I have a brain-powered Roomba? The problem is, biology is messy. The neurons in one study lasted a “whole” 11 weeks! The neurons are living things, so you have to actually keep them alive and stuff. For something like a Roomba, you can just use some code or a simple AI and get the same results with much less effort.
So, why are we talking about neuron-based computing? Firstly, it’s always good to learn more about our brains 🧠. We can take lessons we learn from our brain and use it to design more efficient, brain-inspired computer chips (called neuromorphic chips). We could also learn more about how our brains form and this could give us insights into mental illnesses and brain disorders which could help millions of people. However, our computers and silicon chips are different on a fundamental level. For all we know, there are some things that real neurons can do so much better than anything running on silicon that, for these specific applications, neuron-based chips could be necessary.
So What?
At this point, biological computing probably sounds like something with potential, but a whole lot of problems. The truth is, that’s what every emerging technology is. If the problems were all solved, it wouldn’t be an emerging technology. While biological computing has A LOT of issues, if we figure this out it will change the world. If we could reprogram our cells, we could treat disease at a cellular level. What’s more, if the immense parallel computing power of DNA is realized, we could run problems that we can’t even consider today in a fairly reasonable timeframe (remember, DNA computing does take some time). We probably can’t even imagine the eventual applications of this technology. The first computers took up entire rooms, and now everyone walks around with something unimaginably more powerful in their pocket. We are still in the punch-card era of biological computing, just imagine where we’ll be 50 years down the road.
We’re providing opportunities for the next generation of student thinkers, inventors, and learners, to publish their thoughts, ideas, and innovation through writing.
Our writers span from all areas of topics — from Growth to Tech, all the way to Future and World.
So if you feel like you’re about to jump into a rabbit hole of reading these incredible articles, don’t worry, we feel the same way. ;)
That’s why studentsxstudents is the place for getting your voice heard!
Sounds interesting? Why not join us on this epic journey?
