Atoms, Bits & RNA
DNA is a digital code that runs in the physical world. That puts biology in between two very different ways of thinking. Today we're talking about atoms, bits & RNA.
Transcript
Building technology means living in two worlds: atoms and bits.
The world of atoms is the real, physical world. The one where I have a body, eat hamburgers, and hang out with my friends on a sweet orange couch. The world of bits is what happens inside computers and on the internet. It's software, data science and social media.
Most of us get that this distinction is more conceptual than literal. Computers are physical too, not really magic portals to the cyberspace dimension.
But we shouldn't underestimate how deep the division goes between atoms and bits and how much it shapes our thinking. In Western history, it's at least as old as René Descartes, who described Dualism as the idea that there are two fundamentally different kinds of things: body and mind.
Ever since the birth of the internet, there's been a kind of anti-atom, pro-bit subculture vibe in certain parts of the tech world. In the 1996 Declaration of the Independence of Cyberspace1, it went like this:
Now that's a community that hasn't touched grass since 1996. Just kidding. Love you tech people. And I love bits too. I'm a tech guy.
But I'm also a biology guy. And there is nothing biology loves more than to explode our mental models by throwing out a bunch of exceptions that don't quite fit. So naturally, biology breaks the atoms-bits distinction.
I think this is true for all of biology. But it might be the easiest to see in the case of functional RNA molecules and RNA therapeutics. RNA strips away a lot of the complexity of a typical biological system, letting us see atoms and bits at the same time. An RNA molecule is virtually a naked piece of information that also has very material consequences for human health.
Zoom in on RNA. The sequence is pure bits. It's just a character array coded in base 4. If you email me that data, you and I can synthesize the same molecules that perform the same function, sort of like copying a snippet of computer code.
But the RNA’s function plays out in the physical world according to biochemical rules. An RNA molecule can fold into a 3D structure. In this form, it interacts with other pieces of biology. It can bind to small molecules and act like a sensor. It can catalyze a chemical reaction to act like an enzyme. It can be more or less stable to the effects of temperature, or to the enzymes in the body that degrade RNA.
An RNA molecule can express a protein, if the sequence is readable to the translational machinery of a live cell. For an RNA medicine, this is the therapeutic payload. The bits of the coding sequence determine what is made. Other sequences can determine how much is produced and which cell types produce it.
In order to have these effects, the atoms of the RNA molecule have to be delivered to the right location in the physical world. For a therapeutic, that means inside a human body, where unfortunately you can't send a sequence by email. RNA has to be packaged, for example by wrapping it in a lipid nano-particle. It has to reach the right cell types, interact with the cellular machinery in the right way to produce the therapeutic effect.
If we wanted to summarize this, we might say that an RNA therapeutic represents a sequence in the world of bits, an activity in the world of atoms, and an insanely complicated transfer function connecting the two.
Biology lives in between two worlds and that's what gets me excited. Because sometimes you can get the best of both. The bits-like properties of RNA unlock problem-solving strategies that we usually don't get in the world of atoms.
An RNA sequence is granular, meaning that we can make changes as small as a single base.
It's indexable, meaning we can choose a base for each position independently.
It's homogenous, meaning we always have the same 4 bases to choose from.
It's connected, meaning that any two interesting sequences can be mixed and matched in well defined ways.
Properties like these make it possible to build and test many different versions of an RNA in systematic ways. Learning the transfer function between sequence and activity might be the hardest problem in biotech, but with large and unbiased datasets we can activate machine learning and other powerful digital tools.
Building with bits of RNA is not like building software. But it's not not like software. And it's not like developing conventional therapeutics either. It lives in that awkward in-between space where biology loves to put us.
And I, for one, love it here. I think that this shadow realm that bridges atoms and bits is where the true power lies. When we break the normal categories, it feels a little bit like we broke the game. The world of bits isn't separate from the world of atoms. In biotech, we get to have both. And if we want to do this right, we have to have both.
This was a good read Jake.
keep em coming!!