The following is a guest blog post that I wrote for a friend’s blog (now defunct) in 2009. At the time, I was one year into my PhD and the swine-origin H1N1 influenza virus pandemic was just beginning. I was cleaning through some old files and found it today, and thought it would be an interesting “time capsule” to share with the world. I don’t want to disrupt the flow of the piece, but I’ve added some footnotes that give commentary on what I think of my 11-year-old predictions today.
Swine flu (http://en.wikipedia.org/wiki/2009_swine_flu_outbreak). If you have a television, computer, or live near a news stand, you’ve heard about this new virus. If you’re in the media, you’ve been running the story into the ground. If you’re in the study of biomedical virology, as I am, you’ve had a very different perspective on this virus than most of the general public.
Most of us in science have just been interested in how the virus arose and the genetic lineage it has. Its (pathogenicity http://en.wikipedia.org/wiki/Pathogenicity)—i.e., how severe an illness it causes — was never very high, and as soon as influenza researchers saw its gene sequence, they realized it lacked some of the major markers for virulence. This means scientists have spent most of our time telling people to calm down and go about their lives. A far cry from US Vice President Joe Biden and the media.¹
For me, though, swine-origin H1N1 influenza has driven home another point. It’s actually the media frenzy that did it. I studied the Black Death in college. In that plague, people would flee ahead of it, and the news generally traveled with messengers who were already infected or dying. While swine flu is nowhere near as bad, I did notice something interesting. We now live in a world where news of a disease travels faster than the disease itself.
Really, we’ve lived in that world for more than fifty years now. The Internet has made our media even faster than that; our news now travels faster than our understanding of a pathogen. Disease intelligence resources so far outstrip those of the past that many experts look at swine flu and say “Well, it’s possible this sort of thing happens all the time, and we just never noticed before.”
That means we are on the cusp of a whole new paradigm. We can pass information through the disease intelligence infrastructure from the earliest points in the emergence of a new pathogen.² And that is only the beginning.
There is an effort known as the X Prize for Genomics (http://genomics.xprize.org), which carries with it a ten million dollar award. The goal is to sequence 100 human genomes in ten days, at less than ten thousand dollars per genome.³ Meeting this challenge will change a lot of things about human health, but I’m talking about pathogens, most of which have genomes that are many orders of magnitude smaller than the human genome. When we can sequence ten human genomes in a day, we’ll be able to sequence that many influenza genomes in seconds.
New imaging techniques will also drive our access to information. Instead of needing days of expensive sample preparation and further years of analysis to be able to see our microscopic enemies, new pathogens will be imaged on the first day they arrive in the clinic.⁴ The ready availability of both sequence data and structural data will revolutionize how we interact with infectious disease.⁵
When this day comes, hopefully sometime in the next fifty years, we will be buried in data about the distribution of diseases throughout the world. That data will be pored over by disease specialists, and patterns will emerge that we have not seen before.
What I am getting at is that at some point this century our public health infrastructure will be able to get an idea of what pathogens are circulating in what places in nearly real-time. And more importantly, that infrastructure will know the genetic heritage of circulating pathogens. We will see just where certain traits originate, and where in the world two pathogens with potentially lethal properties will have the opportunity to trade genes.⁶
This has staggering implications. Where in the past we have reacted to new diseases and antibiotic resistances, public health groups like the CDC and WHO will be able to issue advisories to hospitals about what new pathogens are on the horizon. Rather than finding ourselves stuck with antibiotic-resistant “superbugs,” we will find ourselves predicting the emergence of such bacteria. When influenza viruses, for example, coexist in a given population, we will be able to predict the emergence of novel, potentially pandemic strains, and fight them before they appear.
The mountains of data we are about to see on disease intelligence will do something to the fight against infectious disease that no drug will ever do; they will turn our war against pathogens into an offensive battle, rather than a defensive one, for the first time in the history of our species.
¹I think at the time I wrote this, I found Joe Biden to be making needlessly alarmist statements about the severity of the new influenza virus strain. I can’t remember what he said, but now that I’ve seen the alternative — where a public figure tries to downplay the seriousness of a pandemic — I think I prefer Biden’s approach.
²This feels eerie to read now. This is absolutely what happened with SARS-CoV-2. We had detailed sequence information about the virus in the earliest days of the first appearance of COVID-19. The Moderna mRNA vaccine was designed in January 2020. I thought it would take 50 years to get there, as you’ll see later. It took 11.
³Unfortunately the X-Prize for Genomics was canceled, but fortunatelythe pace of sequencing innovation hasn’t slowed.
⁴This one hasn’t quite happened yet. However, we do have amazing new visualization techniques that have proven valuable in the study of SARS-CoV-2 and other pathogens. We’ll get there.
⁵However, we do have amazing capabilities to do sequencing and structural assessment now, much more so than in 2009. Part of this is related to advances in computer-based structure analysis that allows us to predict how a new virus’s proteins may look using structures we have determined for closely related viruses.
⁶We are SO CLOSE to this capability now! The genome of SARS-CoV-2 was sequenced rapidly. Within a month of identification it was clear, based on a history of logged global surveillance, that the virus originated in bats in Asia. The evolutionary history of the virus was determined rapidly and that helped us to understand its heritage and origins. The sequence data that made this possible came from surveillance conducted in the decades since the SARS-CoV-1 epidemic. I was dreaming when I wrote this. It’s very close to real now, and that’s amazing. The only reason I say “close” is that it isn’t quite “real-time.” We aren’t at the point where we sequence everyone who comes in as a means of diagnosis. Sequencing happens when something strange is going on. I think eventually we will sequence the pathogens infecting everyone as a matter of routine care, though.
Overall: I don’t think I would write this piece the same way if I had written it today, but I do think that the past 11 years have really changed the way that disease surveillance is conducted. Still, it wasn’t enough to achieve the predictive capabilities that I envisioned here; if it had been, the emergence of SARS-CoV-2 would have been an interesting footnote in a medical journal instead of a global catastrophe.It’s interesting, though, to see how far we have come. We were able to sequence a totally unknown virus within a month or two of its first emergence. Diagnostics to detect it and vaccines to defeat it were designed around the same time as well. What’s more, the speed of these advances was fast and clearly validated the importance of the scientific process. I expect that the world will be more interested in achieving the ability to predict and contain emerging pathogens than it was before the COVID-19 pandemic. In 2009 I thought that our predictive capabilities would take 50 years to arrive, but I now think it could be substantially sooner.