June 15, 2020

The big thing about seeing small

By Conor McKechnie and Dodi Axelson

The big thing about seeing small

Biology and telecom have more in common than we think. Dodi, Conor, and their guests draw parallels between seeing inside a cell and computing in the telecom industry. Learn what that could mean for our future.

DODI: Conor, I am sharing an image with you, and I want you to tell me what you're looking at.

CONOR: Okay, this is always gonna be good for a radio audience.

DODI: Yeah, but let's bring them along. Now here you are:


CONOR: Oh, wow. Okay, what have we got here?

DODI: Well, you tell me. Describe what you are seeing for our listeners.

CONOR: Well, I don't know. Looks like it could be something from the Hubble Space Telescope. It may be something like the Pillars of Creation, that famous photo looking deep into space into the beginnings of the universe. And it's kind of bright purple, pink, yellows, green. Is it actually a photo? What is it?

DODI: Well, you would be surprised to know it is actually a close up of breast cancer.

CONOR: Okay, so unexpected to see something so dark being so beautiful. When you get really close to it, I guess those are all little cells in a tumor or something like that, right?

DODI: They are.

CONOR: It is extraordinary.

DODI: And beautiful or not, what we're going to look into today is how the smaller you look at things, the bigger the understanding you get.

CONOR: Oh, I like that. And that's what matters on today's episode.

DODI: Yes, sir.

CONOR: So, these are breast cancer cells. What are we looking at exactly? What's the science behind some of this?

GUEST #1: The way we really look at it is by tagging things with fluorescent proteins. And so when you see the pictures, they are beautiful. And depending on what kind of proteins are being expressed, you either get huge swaths of one color or a mosaic of all these different colors. They're really, really beautiful, and each one is unique. So, no two pieces of tissue imaged this way ever looks the same. And that's kind of an indication of the variety of biology that we have inside all of us, some of it helpful and some of it very harmful. But I have to say, as tragic as it can be, you always keep in mind that you're looking at somebody's disease and a life completely altered. They're just lovely.

CONOR: So that's kind of deep. Who's that?

PRACHI: My name is Prachi Bogetto. I am a segment leader in diagnostics.

DODI: And Prachi, who is our colleague, is going to help us look at the miniscule today.

PRACHI: So, I think about diving deep into the cell and all the different ways that technology now allows us to understand a single cell. And the very big thing, from my perspective, that comes from that is there's actually unique discovery happening. We're finding types of cells or cells in states that we had never seen before, and to me that's huge. For all the years that we've been studying biology and cellular biology, to still be in a state of unique discovery is just, it's astonishing to me. I think that's the big thing about looking small.

DODI: You know, for me, when I think of my old science class, I think of firing up that Bunsen burner, dissecting frogs, and of course, looking into that little black microscope on my desk. But Prachi sees things differently at the cellular level.

PRACHI: So, that little black microscope on your school desk still exists. And it's still a very powerful tool. But it's really the grandparent, if you will, of some of the tools that we have now to look deep inside cells. So, if you think about super-resolution microscopy, it allows you to look at an individual live cell. That's really an amazing tool. More than once I've sat at what feels like an ordinary session, showing a researcher what our technologies are capable of. And they'll say, well look at that. I have never seen that before. So, that little desktop microscope that you used is all grown up. And it's driving discovery at the single-cell level. And then if you think about not just looking at cells but trying to understand their function—So why do they look the way they do? Why do they make the things they make?—What you're interested in then is the genetics of an individual cell. And we have many, many different tools to understand the processes and the mechanisms deep inside the cell that wouldn't look like anything you're used to seeing through that microscope, but it's giving just as much information about what's going on inside that cell.

CONOR: Okay, I get it fundamentally. But does Prachi have an example? I want to know what kind of information you can get if you're just looking at such a singular level of one cell.

PRACHI: You're getting information on how that cell is different from cells that look exactly like others under a conventional microscope. So you're getting an understanding of what proteins it is making, because proteins are really what drives all of life.

DODI: So, you see, we're back to proteins again.

CONOR: They do really follow us around on this podcast. And they're like the host, the recurring guest star.

DODI: Yeah, that's right, like New York as a character in a rom com. I heard Prachi mention them before when she said that she tags fluorescent proteins when looking at breast cancer under the microscope.

CONOR: So, what's the cameo for our protein superstar today?

PRACHI: Different proteins are expressed differently in different cells. So you can see that at a single cell level, and then if you want to dig deeper, you can understand why those particular proteins are being made. Why are they being made in that abundance? Why are they being made across a certain timescale? And that really gives you a deep, deep look into the mechanics of that individual cell. And then you can take that and kind of explode it out to so many different things, how cells behave in a normal state, how cells behave in a disease state, and really kind of the holy grail of all this understanding is to say: Can we get to a point where perhaps we can alter that basic mechanism of a cell in a disease state, and instead of treating a disease stopping it from ever starting?

CONOR: So, this takes us towards CRISPR, does it? Gene editing and modifying the instructions for life at a cellular level? Is that what we're talking about?

PRACHI: We could. It might be a short conversation, because my knowledge is kind of skin deep. But if you think about it more broadly than that, how do we deal with disease now, when somebody doesn't feel well, or a test looks a little bit off? You're chasing it after it's already happened in a lot of cases. And if you think about that, at the diseases that we have from birth, right, genetic abnormalities, you're already playing catch-up at the moment a child is born. So, if you could kind of roll back time, if you will, and alter the DNA so that this disease state never happens, then we could talk about gene therapy because really, that's what gene therapy is. That would be amazing.

DODI: So, what Prachi is describing there is single cells and understanding the proteins that a particular single cell is making, and maybe teaching that cell to make a different protein.

CONOR: Or to behave differently then?

DODI: Exactly. Or have a different location.

CONOR: Right, so is that a different outcome in general from gene editing? Or are we looking at something similar?

PRACHI: I think we talk about CRISPR a lot in terms of what can it get us for broad manipulation? So, for instance, there are people right now working on CRISPR-based vaccines for COVID-19. When I think about gene therapy, I think about personalized manipulation, so you wouldn't be looking to solve a really big population-scale problem. Usually with gene therapy, you're going after devastating diseases for sure, but in a smaller subset.

DODI: By understanding a cell at an incredibly detailed minute level, you can tackle broad problems. Prachi says cystic fibrosis is a perfect example of this.

PRACHI: Cystic fibrosis is a devastating genetic disease that typically impacts Caucasian males. You're born with it. Depending on the quality of care and your demographic, you may make it into your early 40's. But it's a very difficult life. And it presents as a tremendous mucus buildup in the lungs, which makes it difficult to fight off life-altering and life-ending pneumonias. What we understand now about cystic fibrosis is, at the cellular level, there's a defect in what's called a sodium-potassium pump. And really, all we need to understand about that is these different elements in our cells really control how those cells behave.

So, patients with cystic fibrosis have, at a cellular level, a genetic abnormality that makes the sodium-potassium pump in cells function improperly. And that is the reason why sufferers have this terrible mucus buildup in their lungs. So, the fact that we know that it's a sodium-potassium pump, and that it’s a genetic abnormality, is a huge discovery. And if we can find a way to alter the genes to get a properly functioning sodium-potassium pump, it changes and saves lives. But we had to dig way deep, way deep down into the cellular level of how ions cross membranes to get that understanding.

CONOR: Okay, this is absolutely fascinating.

DODI: I think so, too. And when I was chatting with Prachi about all of this, our conversation took a turn a bit closer to my comfort zone. You'll recall that I worked for many years in the telecommunications industry. So I asked Prachi if she ever thought about the parallels between science, where you learn so much from a teeny tiny thing, and the information communications technology industry, where massive information is all stored on super tiny chips.

PRACHI: The piece that we don't really talk about when we talk about this ability to look at cells in this kind of detail is the computing power that all of that data feeds. And if you're feeding all of this data into computers, which are essentially electronics, the smaller and faster they get, the more and more information they can process. But the more and more data they need. They're completely symbiotic; one doesn't exist without the other. We don't talk about that very much because we're so fascinated by what happens inside the lab. But once you get that data, that's where kind of the movement in computing power towards smaller and faster really comes together.

DODI: And this in turn led us into the world of Moore's Law. And this guy.

GUEST #2: Hello, my name is Cristian Norlin, and I work as the head of design technology at Ericsson One, looking for the game changes that are going to propel Ericsson into the future.

CONOR: Okay, so Moore's Law. I mean, we're getting towards the edges of Moore's Law really aren't we? But why don't we remind people what Moore's Law is all about?

CRISTIAN: Well, it's about exponential growth, basically. You can say, in a nutshell, it's almost like a formula that actually held true for a number of decades where the projection was that a tiny amount of increase in solid electronics created this massive amount of effect over time. And, actually, it really doubled the size, whatever it was. It actually made sense for a long time, which was kind of strange, almost like a self-fulfilling prophecy in a way that almost felt like the industry just adjusted to that law saying, "Alright, so this is the base we're going to go for."

DODI: So, what that resulted in was basically having a lot of computing power in your pocket. And it's become a cliche now that all of the computing power in our pocket was enough to take people to the moon, right? We have more power in our iPhone™ than NASA had when they sent Buzz Aldrin and everyone else to the moon in the 60's.

CONOR: Yeah. And we use it to share pictures of cats. It's brilliant.

DODI: Isn't it, just?

CRISTIAN: Oh, yeah, I mean, everything basically that we can see today is the result of that. It's almost like a law of nature that from the early days of computing to where we are today has actually followed that trajectory, that prediction, rather well. Today, you can say that it's becoming slightly more complex in terms of when you talk about information density, you can see that we are not confined when it comes to computing power and the way we analyze things and do things. We're not confined to one single processor; we're actually starting to see the benefit effects of scale. For instance, if you're talking about massive computation today, it's not happening in one cluster. It could actually be distributed across numerous chains and clusters that make the Moore's Law kind of thing more difficult to solve, to agree on.

CONOR: So, for the second time, I'm going to ask for an example. Does Cristian have something that explains this?

DODI: Of course he does. And that is something called mixed reality.

CRISTIAN: Mixed reality is basically when you take virtual objects, and you blend them with the real world. You have probably seen the Magic Leap goggles, where you look through a pair of goggles, and you can see virtual objects mixing with the reality that you see in front of you. A lot of that sort of computation you can see in the mobile phone today, and Apple and Google have those kind of AR apps and functionalities. Everything today happens in that phone. All the computing happens there. What's going to happen over time, and this is not even a prediction, it's happening in real time, is that you are getting to a level where you want to do highly complex calculations. And for better processing reasons, you cannot do that on a mobile phone, the technology isn't there, you don't have that sort of processing.

The way to solve that is to distribute the processing from the mobile phone to, in this case, you can place a processor in a base station in an antenna from a mobile network, which is basically a very short distance. So you offload computation. And then you can offload even more computation further back onto centralized data centers with supercomputers and stuff. So you can see that the computing that you experience when you have your phone or your goggles on is actually happening across a chain of different devices. And I think that's where Moore's Law becomes more difficult to say, all right, yeah, it works. But it's still sort of there anyway.

CONOR: So, this now brings us back to the biotech industry.

DODI: Yeah, Cristian told me that he thinks his industry, information communications technology, will learn more and borrow more from the biotech industry going forward.

CRISTIAN: The way you think about technology development is always that it has to do with engineering, or algorithms, mathematics, physics, whatever. But if you think about it, nature is among the most complicated organisms and systems. There is actually a rather extensive body of research and thoughts: What if we can actually learn from how nature has organized itself, how nature solves problems and apply that to new technologies as well?

If you think of solar cells, that's an interesting case. One problem that they have is that they get really warm from the sunshine from the energy from the sun, and then they perform far worse. You can look at, for instance, butterflies, which is interesting in nature to have among the best capabilities to absorb heat. And I mean, they have really thin wings, but they do not burn. They channel that heat and it actually has to do with biology. So the solar cell industry has looked into that and said, "Well, that's interesting. What if we do something that mimics that?" And I think we can see a lot more of how nature organizes itself, the seeming chaos in organic systems that actually turn out to be really valuable.

CONOR: So, this is absolutely fascinating. And we're seeing this in lots of different areas where biological computing is potentially becoming a thing. And I'm kind of excited because we've been talking about the combination of biology and engineering and computing being kind of the 21st century's big leap forward. And this seems to be exactly where Cristian says we're going.

DODI: That's right. All the lines are blurring. There used to be separate industries. But the truth is, we can all learn from each other, can't we? Because every day is a school day. There we have it. And that's it for this episode. Thank you for listening to Discovery Matters.

CONOR: And rate us on your favorite podcast app. Thank you. We'll see you next time.

See more images captured by Cell DIVE imaging.

Listen to more podcast episodes.