Alter Everything Podcast

A podcast about data science and analytics culture.
Alter Everything Podcast Survey

Be entered to win a prize while sharing your thoughts about our podcast.

Give your two cents here
Note: Alteryx Community may experience brief downtime during 12a-4a PT hours. We apologize for any inconvenience.
Alteryx Community Team
Alteryx Community Team

Alteryx CDAO, Alan Jacobson, had a chat with author and scientific journalist David Quammen about COVID-19 to discuss the origins of the virus, where we stand historically against other pandemics, and how we can use data to work together towards eradication.

Special thanks to JW Legacy Music, (@jesperwinkelhorst) for the awesome theme music track for this episode.




Alan Jacobson - @AJacobson, LinkedIn, Twitter
David Quammen - Personal website, LinkedIn, Twitter



Copy of Podcast for slides.png





MADDIE: 00:03

This is Alter Everything, a podcast about data science and analytics culture. Today Alteryx Chief Data and Analytics Officer, Alan Jacobson, is our guest host. You've heard him on the podcast before. He sat down with author David Quammen, a prolific writer of 16 books. One very notable given our current challenges is his 2012 book called Spillover: Animal Infections and the Next Human Pandemic, in which he focuses on how diseases like the coronavirus spread. In David's research, he really knows a lot about ways that we've dealt with pandemics like this in the past, and he draws on that knowledge to help find parallels between what was done at the past and what we can do better this time. This episode is a great reminder of how analysts and data scientists are primed to really help be a part of the solution amidst global uncertainty. We often talk about how analysts are the unsung heroes, and after his conversation, Alan reflected on so many different morsels that are helpful to keep in mind when trying to be a part of the solution. Hopefully, after this episode, you feel a little bit more optimistic and maybe even motivated to find unique ways of using your analytics expertise to be a part of the solution. Let's get started.

ALAN: 01:21

Our guest today is David Quammen. He comes to us today from his home in Bozeman, Montana. Welcome, David.

DAVID: 01:27

Thank you, Alan, good to be with you.

ALAN: 01:29

So we're in quite a period of time with a pandemic that's clearly affecting businesses and lives all over the world. So first, how are you doing?

DAVID: 01:37

I'm doing fine. How are you doing? How is everybody doing?

ALAN: 01:41

Yeah, the new normal is taking a little while to feel normal.

DAVID: 01:45

Yes, it is.

ALAN: 01:46

So one of your core areas of research and writing is around diseases that spread from animals to humans. Can you tell us a little bit more about how this happens?

DAVID: 01:55

Right. So the book Spillover is about zoonotic diseases. Zoonosis is an animal infection that's transmissible to humans. It might be a virus, it might be a bacteria, it might be any other sort of tiny infectious bug. If it transmits into humans and causes disease, we call that a zoonotic disease. 60% of the infectious diseases known among humans fall in this category. In the strict sense, they are in recent terms zoonotic disease resulting from this crossover of infection from non-human animals to humans. How does it happen? Well, it happens by close contact between humans and generally wild animals. Close contact might involve hunting a wild animal, killing it, butchering it, getting blood in a cut or simply from eating it. Or it might be not quite as intimate as that, but coming in contact with a wild animal because it's caged in a market where you're selling seafood or where you're buying pork and the virus or whatever else is the agent might have spilled out from that wild animal and gotten into that other animal, which we would call an amplifier host. And the virus might have become amplified in that host and then passed on to any number of humans. That's probably what happened in China.

ALAN: 03:21

Right. So one of the things I've seen is that this virus has been called slippery and that it's already mutated. I think there are two known strains already. Does this make it normal or different from the norm? Should we think of it differently because of this?

DAVID: 03:36

This virus, because it's a coronavirus, belongs in a category of viruses, a family of viruses that are known to have high intrinsic evolvability, is what the fancy term is for it. They evolve quickly. Coronaviruses are known for that and they've been known for that for decades, which is why this thing being sort of taken by surprise is so frustrating. Why do they evolve quickly? Well, coronaviruses and some other families of viruses have a certain kind of genome. And again, the fancy terminology is it's a single-stranded RNA genome, as opposed to perhaps a DNA genome where the molecule is the famous double helix. Double helix DNA has more stability. Single-stranded RNA, which carries genetic information also, is unstable so that when it copies itself as the virus replicates, one virus getting itself copied through the machinery of the cell that it has infected, getting itself copied multiple times and eventually bursting out of that cell as many virus particles, the copying is inaccurate. There are mistakes made. That results in genetic mutations, and those genetic mutations represent variation within the population of viruses. The virus particles are competing against one another under circumstances of genetic variation. You turn the crank, and you have evolution by natural selection. Those are the exact conditions that Darwin talked about. Variation within a population, competition, differential reproduction, natural selection, adaptation, evolution. That's what's happening with this coronavirus.

ALAN: 05:28

So can you give us some perspective on what you're seeing with COVID-19 relative to other pandemics and other coronaviruses?

DAVID: 05:36

Sure. Well, the most famous of other Coronaviruses was the SARS virus. In 2002, a new virus got into people in southern China near the city of Shenzhen, which is not too far west of Hong Kong. How did it get into people? Probably coming from a bat and going through an amplifier host. And this virus, which we now call the SARS virus, started spreading quickly and it got to Hong Kong. Eventually, it was stopped - the SARS outbreak - after "only" about 8000 people got infected, and about 770 people died for a case fatality rate of almost 10%. That was a coronavirus. That told scientists who were paying attention that coronaviruses could be very dangerous. And they have known that. And I listened to them when I was researching my book Spillover, so I have known that and other people have known that.

ALAN: 06:44

So at Alteryx we're obviously focused on analytics, and one of the things we've been seeing is how the use of data and analytics are becoming more and more common in efforts to combat disease and disease spread. We see it with cell phones being used to track the epidemiology of the spread, we see it being used to predict where there might be risk for future outbreaks, for proactive quarantining. We've seen analytics being used to identify potential treatments and hackathons - open-source hackathons - to find new drugs, and even diagnose people with AI interpreting lung CT scans. I'm curious if you're seeing this trend with analytics becoming more a part of the equation? And if so, do you have any sense where the technology might be used next?

DAVID: 07:34

Well, absolutely. Certainly, molecular biology, molecular phylogenetics, molecular diagnostics are extremely important in this situation. First of all, when the genome was sequenced from this virus early on by Chinese scientists at the Wuhan Institute of Virology, that gave important information that said right away, "Bingo. This thing is a coronavirus." What else did that genome sequence say? Well, it told people that okay, this is related to SARS because they're both coronaviruses, but this is different from SARS. Different enough to be considered a different virus. However, it's closely related to that other coronavirus that we found in horseshoe bats in the province of Yunnan about five years ago. The same people with some international collaborators and co-authors on the paper. So they said, "Okay. It's very, very similar. I think 98%, 99% genomically similar to this virus that we found in horseshoe bats in a cave in Yunnan. So the chances are that this thing is that virus with a little bit of mutation, a little bit of evolution, and it probably has come from a horseshoe bat." That's important. It's not important in terms of how to control it, but it's important in terms of lessons to be learned and how to deal with the next one.

ALAN : 09:09

So interesting. Thinking about the scientists researching the evolution of genome sequences and viruses, I can't help but think of data scientists who study historical data to make predictions for the future. What these scientists are doing is advanced analytics.

DAVID: 09:26

And then, as you said, there are, I think it's two major strains of this virus now. I saw a tree of life of this virus. I need to look back at that, but my recollection is that they can see now from sequencing all our parts of the genomes of the virus as it appears in different human patients around the world, they can create a tree of life. Molecular phylogenetics, which happens to be the subject of my most recent book-- not Spillover, but a different one called The Tangled Tree. So molecular phylogenetics of the new coronavirus tells us how quickly it's evolving. It tells some things about how transmission occurred. For instance, if I recall correctly, and don't hold me to this, that one of the major limbs on this COVID-- I shouldn't call it COVID-19 tree of life because COVID-19 to speak accurately is the disease, and SARS-CoV-2 is the current name for the virus. Anyway, the tree of life of this virus shows that one of the major limbs is the one that got into South Korea pretty early on and apparently hasn't interacted spread from there and not too much. Well, it hasn't spread from there as far as has been identified. So you can learn some things about where things came from.

ALAN: 10:52

Where things came from. I can think of so many examples for why it's important to keep track of where things came from when doing analysis. From a basic level of making sure your data is clean to a more advanced level when doing spatial analysis, say, in tracking geographic patterns of movement of a virus.

DAVID: 11:12

For instance, you could sequence the genome of people in Milan who are infected or have been infected with this virus, compare that to other genomes, and you could perhaps figure out how this severe outbreak in northern Italy and now in all of Italy, how that occurred. Did that come from straight from Wuhan, China? Did it come from somewhere else? Was it a worker or a businessman or a tourist? How did it get there? You could learn all these things from what I would call molecular diagnostics or molecular phylogenetics.

ALAN: 11:51

Yeah. I mean, when we think of analytics, it's simply using data, finding patterns in data, analyzing data to get insight.

DAVID: 11:58

There we go. Okay. We're talking about the same thing.

ALAN: 12:01

Yeah. Yeah. And I would actually suggest that in much of your writing, you are actually doing analytics. It might be qualitative in nature, sometimes quantitative but frequently qualitative in nature. But you're looking at the history and the patterns of data and forming insights as to what might happen next.

DAVID: 12:20

Absolutely. Absolutely.

ALAN: 12:22

Analytics is really just taking data and getting insight from it. Sometimes the data is quantitative, bunches of numbers, but many times it's qualitative as well. David mentioned interviewing people and analyzing the patterns, and this is also a fundamental piece of what analytics is all about. Whether it's done manually, as many of David's example shows, or done with computers, it's still great analysis and analytic science.

And so with that, we normally see domain experts: the microbiologist, the epidemiologist, as typically being the superhero at the end of the day. It's usually and frequently not actually the data scientist. Data scientists frequently help, but there's usually a domain expert that's at the core of coming up with solutions. Accountants coming up with new ways of doing accounting, lawyers implementing natural language processing techniques to change the way litigation's practiced. But at times, there are ways that people outside of a domain can jump in and find new ways to help solve a problem or transform an industry outside of the experts. I'm curious, in this arena, if you see some solutions being offered up with players who are not the normal established players in the space - the healthcare researchers, the drug companies, the myriad of professionals that are focused on this every day - or if you see it mainly coming from within those domain experts.

DAVID: 13:52

Well, that's a good question. Let me think about that. And let me start talking about what I know and see if that leads us to an answer. One of the things that I see in this kind of work - and I described this in Spillover - is a partnership of a sort between field disease detectives, laboratory people, and I suppose analyzers. For instance, say there's an outbreak, a new outbreak of some kind of a virus in Malaysia, people are getting sick--

ALAN : 14:25

David took us through a great example of how mysterious diseases can start, and how they can easily appear to be very similar to others. And when this happens in a different area of the globe, maybe they spread to other locations quickly, or maybe different strains pop up simultaneously. Good communication becomes paramount, certainly something we see in data science.

ALAN: 14:50

It's really, really fascinating in that the number of different experts in different domains, each knowing a piece--

DAVID: 14:56

Yes, exactly.

ALAN: 14:57

--of the mystery, but none of them by themselves maybe having the entire answer and potentially could be misconstruing the answer had they not been able to get to another person who could add a little more context to the story. So certainly a diverse team of individuals versus any one individual, which frequently we see that in analytics, that it's domain experts sometimes paired with data scientists and people with other skills that frequently can make big discoveries and make big transformations happen. So that part seems a fairly familiar story. I have noticed that obviously, as this virus hit, the number of non-experts who wrote about it certainly went up dramatically. And I'm curious your views on that because obviously, there's both information and misinformation now that is out there. And are there ways to help mitigate the misinformation?

DAVID: 15:56

Yeah. Well, there are. I mean, everybody's writing about it. Everybody's talking about it, including a lot of non-experts. I consider myself a non-expert. I'm a fellow who listens very carefully in a great length to experts, but I'm not an expert. But I've been on radio, television, writing op-eds, etc, trying to be very careful sticking to the facts. And in my case, I try and stick to what I know, which is the ecology and the evolutionary biology of these things. I try not to say too much about public health measures - what should we do now - except everybody wants to know that so they ask you those questions. I listen to [Dr. Anthony] “Tony” Fauci because I trust him. So you listen to experts that you trust, you know, and you trust.

ALAN: 16:44

When analyzing any situation, any piece of information, or in our case, any data set, understanding the source of information is so critically important.

DAVID: 16:54

And in my field, you trace back to the actual science as it's published in legitimate scientific journals, and you certainly don't trust something just because somebody is saying it online. It's a broader problem, as you well know, Alan, in our society. How do you tell what's real on the web? You need methods, you need critical thinking, you need some standards, because somebody is saying every conceivable thing on the web, and a very great amount if it is not true.

ALAN: 17:41

So as someone who's clearly done a lot of research on many other pandemics and events like this that have happened in the past, what's your perspective of kind of where we're at and what the near term and longer-term future holds?

DAVID: 17:57

I think we're in the middle of a long series of outbreaks that might each become a pandemic. Stretching from the past, from about 1961 - at least I can identify a starting point - not just up to the present, but into the future. Because once we get this thing under control and get the fire put out, immediately, we should start planning for the next fire. And we should be much more ready for the next one than we have been for this one.

ALAN: 18:30

Yeah. So it certainly seems that the data would suggest this one won't be the last. And as you said, being more prepared and as the world gets effectively smaller and smaller as we are better able to travel and people are able to move around the globe, more and more freely.

DAVID: 18:49

Oh. I guess just one thing that we haven't quite touched on. Besides there being 7.7 billion of us, the fact is we're disturbing wild, diverse ecosystems, more and more. And each of those ecosystems contains great diversity of animals, and that diversity of animals contains a huge diversity of viruses. So the more we disturb wild ecosystems, the more we invite viruses into us. I know some wonderful virologists who study the diversity of viruses. Forest Rohwer studying the diversity of marine viruses. There's a fellow named Trevor Bedford. There's a fellow named Eddie Holmes in Australia. They're doing the viral discovery and the analytics to tell us just what is out there, and what might be our next problem. And it's great important work.

ALAN : 19:48

How crucial that we all share our ideas in a positive and productive manner in order to uncover solutions for challenges of great magnitude such as this, assembling diverse teams and encouraging people to ask questions, to speak up, and contribute. Once teams are assembled with ideas at the ready, data science and analytics come into play. Our CEO at Alteryx, Dean Stoecker, published a piece in Forbes called “Analytics Against COVID-19: Creating Stability In A Time Of Turbulence.” In the column, Dean labeled analytics a stabilizer, explaining further that quote, “analytics can serve as a stabilizer. Whether you're a data worker in healthcare tasked with optimizing hospital capacity and transportation tasked with reevaluating flight schedules or manufacturing tasked with determining supplier capacity, analytics can serve as both a trusted adviser and your most powerful defense when making a decision that has significant consequences. Data analytics, like many scientific practices, is often associated with fact, logic, and precision rather than emotion and yet it informs human decisions that impact personal outcomes.” Dean also calls on all of us to react with resilience and to be a part of the solution. How will you be part of the solution? How will analytics be part of your positive contribution? Whether the stressors of our current challenges or the next one that comes, share your ideas with us on Twitter using #altereverythingpodcast or comment in the show notes at Thank you for listening.

This episode of Alter Everything was produced by Maddie Johannsen (@MaddieJ).

Alteryx Alumni (Retired)

Very interesting and timely episode!