Dr. Jonathan Eisen is a professor at the University of California Davis, with appointments in the Department of Evolution on Ecology and the Department of Medical, Microbiology and Immunology at the medical school as well as the Genome Center. He's the director of the UC Davis Microbiome Special Research Program and he is an elected member of the American Academy of Microbiology. For his research, Dr. Eisen applies genomic tools to document and make sense of the vast diversity of microbes living on diverse environments on Earth, including beneficial and pathogenic microbes, living on us. Dr. Eisen has pioneered the use of genomic sequencing tools and computational methods to facilitate the discovery of novel pathogens like the one that we are dealing with right now, COVID-19 pandemic. Dr. Eisen is also a leader in the Open Access Movement of scientific publishing, which has been really critical during the COVID-19 pandemic as researchers around the world are rushing to produce, ensure, and analyze scientific data and try to make sense of that data and try to make quick decisions.
*The introduction above was done by Dr. Santiago Ramírez of UC Davis College of Biological Sciences.
Dr. Eisen's Town Hall video with English and Spanish subtitles can be found here.
Transcript of Dr. Jonathan Eisen’s talk
I've spent the last 30 years, which is scary, working on using DNA sequencing to study microbes.
I started studying these weird bacteria that lived inside organisms from the bottom of the ocean or from sediments, like giant tube worms that have strange bacteria that live inside of them. And at that time, it was difficult and expensive to do DNA sequencing, and so I spent a year and a half working in this lab. I worked as a technician for a year after I graduated and I generated data for a single gene from one bacterial symbiont that lived inside one clam species. Actually, one single clam sample from Woods Hole, Massachusetts from the seagrass beds, and it took me a year and a half to read 1,500 or so bases of DNA sequence from this one bacteria and I got a paper out of this. My first scientific paper studying bacteria through DNA sequencing. And today in my lab, undergraduates and graduate students can read 50 billion bases of sequence in a day or two for $1,000. And everywhere in between that time, 30 years ago and now, I've been working on using DNA sequencing to study various microbes, either microbes that we can grow in the laboratory or microbes from environmental samples. And I've been all over the map, I like everything, I can't pick and choose, for better or worse.
I moved to Davis about 15 years ago and still doing DNA sequencing but most of the work we've been doing in the last 15 years has been using sequencing to study communities of microbes, the so-called microbiome, such as the microbes that live on plant roots or on skin or in the human digestive tract or in soil or in other places. And these cases, we're doing this, in part, because those communities are interesting, but also, it's the first time we've been able to study a lot of these communities because DNA sequencing is cheap enough to study thousands of organisms at one time, basically. And so, I have done a lot of work on emerging infectious diseases over the years.
I worked and did the DNA sequencing for the anthrax bioterrorism incidents that happened right after September 11th and we were involved in working with the FBI on doing forensic investigation related to anthrax from letters and from other environmental samples. But I'm interested in everything about microbes, including emerging infectious diseases. And so when the pandemic started, we've just shifted a lot of the things that we're focusing on both in terms of communication and starting to do some research in this area trying to contribute to studies of how the new coronavirus spreads in particular environments, how viable it is in surfaces, whether or not you can get infected by touching particular surfaces and also trying to just help everybody at Davis and in California figure out ways that we can contribute to studying this emerging infectious disease. And since most of what I know is about DNA sequencing, that's the part that I'm trying to contribute to. So I'll leave it there. Basically, if it's about DNA sequencing, I know something about it. If it's about bird photography, I'm happy to talk about that or Open Science or anything in between.
Transcript of student questions and highlights
Q: How important it is to understand and catalog that vast diversity of microbes to predict the next pandemic? And is that even possible? Can we, by knowing every single microbe that lives on Earth, can we make a prediction of when the next pandemic is gonna be?
I think, there are actually some groups at UC Davis that have been involved in this project called Predict that have been trying to do exactly this type of thing with pathogens that probably know more about this than I do. But I have a component of this that I've been a bit obsessed with for the last 15 years or so or maybe even longer…
I wanna be able to know what the distribution pattern is for all microbes on the planet over time, and how that changes in relation to environmental conditions and in essence, build a field guide to the microbes, just like we have a field guide to the bird and the butterflies and the orchids and any other type of organism and it's sort of a crazy dream because there probably are hundreds of millions of species of microbes on the planet. And therefore building an actual field guide that covers all of them is really basically impossible. But the concept of the field guide is a really important goal to have for microbial diversity and it relates to Santiago which is if we want to rapidly respond to novel microbes that are showing up in the wrong place and particular ones that might be responsible for an emerging infectious disease, we have to know what normal is, we have to know what the normal patterns are, of the expectations of where microbes should be, we have to know how to identify new microbes even if we haven't seen exactly that time previously, we have to be able to know something about their biology. So I think building a global catalog of microbial diversity is critically important for being able to respond, not just to emerging infectious diseases, but climate change, and any other event that happens on the planet related to microbes. So I've been sort of pushing this idea maybe for 15, 20 years, and not specifically focused on coronaviruses or anything like that. But I think it is of particular importance for emerging infectious diseases to really have an understanding of the vast diversity of microbes that are present on the planet, so that if something changes, if one of them shows up in a new place or in the wrong place or at the wrong time, we can respond and figure it out.
Q: You mentioned your involvement in maybe one of the greatest detective stories of modern history, which was after 9/11, for students who don't know, about one week after 9/11, there were, sent out to media companies, to congressional offices and it was a huge mystery where these were coming from. You were involved in this and I wondered if you could just enlighten us on how, at the time, modern genomics helped to solve this mystery.
Yeah, I mean, it was a crazy time. I think it was more like two weeks maybe after September 11th or something to that effect. And actually, travel had just sort of restarted after September 11th, and I was in Yellowstone National Park at a conference. So anyway, at that time, this was obviously a very stressful incident on top of September 11th, to have a biological weapons attack within the United States and letters sent to various places. The reason that the place that I was working at got involved in this was a very forward thinking program officer at the Office of Naval Research before had gotten us a contract to sequence the first anthrax genome. And so we had been working for about two years on generating the genome sequence of one strain of anthrax. So anthrax is caused by a bacterium, Bacillus anthracis. It's closely related to a model organism, Bacillus subtilis, but not that many people were working on at a time, but he thought that it was going to be important to generate data from organisms that might be related to biological weapons attacks. And so we were working on this one particular strain of anthrax known as the Ames strain, that was a model organism version of anthrax. And when the letters happened, they first went to a scientist in Arizona to help them characterize the type of anthrax found in the letters. This is a guy named Paul Keim, who had been doing PCR-based surveys of anthrax across the globe. And he showed very rapidly that the anthrax in the letters was basically incredibly closely related to the Ames strain of anthrax. So here we were working on sequencing the genome of the Ames strain of anthrax, and at the same time, the Ames strain of anthrax was in these letters. And so very rapidly, the FBI and the NIH contacted our institute and developed a contract for us for genome sequencing of all of the different samples that had been obtained from the bioterrorism incident. And so, this was, there were hundreds of law enforcement officers working on various aspects of this case, all sorts of different pieces of evidence. And what we can do was generate genomic data that might, in theory, be useful for doing the, in essence, the forensics associated with this investigation. And what ended up happening was both interesting and a little bit scary. What we rapidly realized was the anthrax in these letters was, in essence, identical to the Ames strain of anthrax. And what this meant was, I mean, it had a few differences that I'll talk about in one second, but what this meant was that someone got access to the Ames strain of anthrax and used it for these letters. And, unfortunately, thousands of people had access to the Ames strain of anthrax at the time because researchers used it across the globe to study the biology of anthrax and it wasn't trivial to get access to it, but probably thousands of people had access to it. And what the FBI wanted to do was to try and figure out if there were subtle differences between different versions of the Ames strain of anthrax. So imagine, in essence, think of it like someone gives you a sourdough starter culture, and if they give you a sourdough starter culture, and you then transfer the sourdough starter culture to a friend of yours, it's very, very similar to the sourdough starter culture that you had. But maybe there's been one or two mutations in one of the bacteria in the sourdough starter culture that you can tell the difference between your version of the sourdough and someone else's version of the sourdough. And that requires a lot of work because the genome of any bacterium is millions of bases long and if there's one single mutation and that has happened in the millions of bases, that's pretty defined, but they gave us a lot of money to do this work, and so our institute spent about six months or a year generating genome data from all of these different samples of anthrax and showed that you could, in fact, do even finer scale resolution and identify which version of the Ames strain of anthrax was associated with these letters. In essence, it was epidemiology going back to your original question, tracking the spread of the Ames strain of anthrax among different places. And I could tell you much more about it. It became even more complicated investigation, but basically, the DNA sequencing helped narrow down the possible sources of the anthrax used in the letters, for the terrorist incident. And it's no different than what people are doing right now with what's called genomic epidemiology. If you have an antibiotic resistant bacterial infection spreading in a hospital, you can use the same methods to try and determine who gave that infection to someone else. And you can do the same thing, in fact, there is the same thing with coronavirus. And actually, yesterday, we had a discussion with people in the hospital at UC Davis about possibly working with them on trying to do some epidemiology factions within the health system by doing genome sequencing.
Q: One of the questions was on viral genomes showing the COVID or the SARS-CoV-2 showing surprisingly low rates of mutation given this global spread. Could you comment on the reasons for this genetic stability and how it's contributed to the spread of the virus around the world? We talked a little bit about this last time, but it's an obvious point of interest for students.
Every organism has its own mechanism of copying its genome, and many RNA viruses have error-prone methods to copy their genome, and therefore, their rate of change per generation is pretty high, so HIV is a good example of this. It just doesn't copy its genome very, very accurately in each round of replication, it doesn't. Most organisms like bacteria in humans, when they copy the genome, if they make a mistake, there are error correction mechanisms that come in and fix those mistakes. Some organisms do this really well and others don't do it really well. It turns out, at least in the studies that have been published so far for the new coronavirus, it looks like the error rate in replication is not exceptionally high. Certainly compared to some other RNA viruses, it is much better at copying its genome than many of those other RNA viruses. And that means basically that probability of change, if nothing else happens, is certainly lower than you would expect HIV, for example, but it's changing. It's certainly, I mean, if you look at the gene, there are now thousands of genomes available for this new coronavirus. They are being shared in an open manner to a database anybody can download and look at and analyze. That's one of the things that was developed for the flu pandemic about 10 years ago that is now being used extensively for the coronavirus and what people have found is that the rate of change is not exceptionally high. I mean, it is changing, but it's sort of a medium evolutionary rate. But it's either under incredibly strong selection to prevent changes from happening, which may be part of the story, or the rate of mistakes in the copying of the genome is not exceptionally high. Either way, it's not changing at an extraordinary rate, which is a good thing for many reasons. That means that vaccines that are targeting particular regions of the genome of this organism or particular proteins or parts of the organism are maybe more likely to work for a longer period of time because they may not be changing that fast.
Q: So kind of related to some of that discussion, there was a question about a conspiracy theory that was out there for a while that SARS-CoV-2 was bioengineered. And I know there's a paper that came out recently suggesting that was, in fact, not the case. But the question was, what were the tools that can be used to investigate this? Are there certain characteristics of a bioengineered virus? Based on what you're telling us, how would you find it?
So that's a really great question because about 15 years ago, right before I moved to Davis, when I wrote my first paper, using a field guide to the microbes, I also wrote a paper at the same time that I sent to the same group, DARPA, about how to detect genetically engineered microbes that might be related to a biological weapons attack.
Basically, the idea of this relates directly to the field guide to the microbes, which is you need to understand the natural genetic diversity within a population of organisms. And you need to understand how people might go about genetically engineering microbes and then what you do is you look for whether or not the patterns of diversity in a particular microbe are consistent with natural diversity or if they're really far away from the natural diversity, you look for signatures of genetic engineering. We actually showed that you could do this and we showed that you could detect it for known cases of genetically engineered biological weapons. I think that's basically what people did. We didn't do that for this case. We haven't worked on this for about 10 years but people basically did that exact type of analysis in that paper that you mentioned and they showed that the patterns of variation in the novel coronavirus are completely consistent with it being purely natural and no, absolutely no signal whatsoever that there was any attempt at engineering.
Q: Given that there's variation between different coronaviruses and then as this particular coronavirus expands, how do we begin to tie those changes to understanding viral gene function or changes in viral gene function that could be connected to the clinical side of what this virus is doing. Is that something that's easy to do from a lot of sequencing? Or it's gonna require a whole other set of scientists looking at host interactions?
No, no, I mean, you can do both. So I mean, I, this is what I worked on for many years at this big genome institute. So back then we didn't have thousands of genomes for particular organisms. But we, for example, got 20 genomes for streptococcus and worked with a company called CHYRON to help them design vaccines based upon analyzing those genomes and looking for particular patterns of variation in the genome that would be consistent with the immune system, recognizing those particular parts of the organism, and the organisms evolving in particular ways. So you can scan through genome data with certain models and ask questions about, are there any regions of the genome that are evolving in unusual ways that are suggestive of some type of unusual selection being applied on the organism, for example, for variation in surface protein so that's what we were doing in streptococcus but you can do the same thing with the new coronavirus. So there are many I mean, there's papers being posted every day where people are doing exactly that for the new coronavirus. It is true that you can do that but getting clinical data to relate to that or getting any other type of environmental data about virulence or severity of symptoms or other types of clinical information can make that type of analysis much, much better. And so there are a huge number of people, many of them, some at UC Davis, many at UCSF, many around the country that are trying to take thousands of genomes that have been released for the novel coronavirus and integrate that with clinical data associated with the samples from which those genomes came to try and answer exactly the question that you asked, which is, are there any features in this that can tell us something about the function that we don't yet know, like, for example, if they I'll just tell you like, I don't know this for the coronavirus, but in other bacteria, there are certain types of changes and some surface associated genes that allow the organisms to avoid the immune system. I'm sure that that type of thing is gonna be going on in this coronavirus. I haven't looked at the papers on it but that's the type of thing that you can do with integrating genomic data with clinical data.
Q: There was a related question and maybe you could expand a little bit on how we use that kind of information to generate or target vaccines or proteins for vaccines. I think the question had to do with how are vaccines made really. We haven't gotten to that in the course yet, the idea that we could use dead viruses or we could actually target specific surface proteins or know which surface proteins were changing to help us design vaccines? Is that something that you could fill us in a little bit about?
I don't work right now on anything to do with vaccine design, but I have in the past worked on the computational analysis that would identify candidates, regions of the genome of an organism that might be of interest. And so, what we did is basically the same type of thing that you could do in this case, which is you can ask, so for the novel coronavirus, you can ask, are there any regions of any of the genes that are evolving in such a way that suggests that they are very constrained? So if they are very constrained, that is they can't change much over time, that can be a good target for a vaccine development because then if you target that region of the coronavirus for a vaccine, it's more likely to be effective over time because it's a region of the genome of the organism that is, the organism has a hard time changing. So that's one way that people design vaccine targets. But another way is that it's important is you don't wanna just target a region of the genome that doesn't change. You wanna make sure that you're targeting a part of the organism that is visible to the immune system of the host that you care about, in this case, humans. So if you stimulate the immune system of humans to recognize a part of the coronavirus that is never exposed to the human immune system, it's not gonna help you 'cause then when you get infected with a new coronavirus, your immune system will never see that part of the virus. So you have to balance the goal of trying to target conserved regions, we're trying to target regions that the immune system will recognize when you have a new infection.
Q: So commenting on how open access journals can make a difference in communicating science during a pandemic. Is there a middle ground to be found where both sides open access and private industries can benefit? What's the biggest hurdle and open access?
Yes, so I've been heavily involved in what could be broadly called the Open Science Movement. And one component of Open Science is open access to the scientific literature and the other component is open access to scientific data. Another component is open source software and open access to sort of tools, and there's all sorts of different components to Open Science, and there are many benefits to openness, there are some drawbacks and negative aspects of openness. And finding that right balance is always a challenge. I just wanna mention something related to the anthrax case, which I always find very interesting. When we were doing the work on anthrax, we were a little bit worried that we weren't gonna be, people weren't gonna be allowed to publish the results because it was this criminal investigation and of all the people I was so surprised, the FBI was most in favor of all of the data being released as openly as possible. And they have this great argument which was they didn't have all the experts in the world that knew how to work on anthrax so they wanted help from other people. And the more that the data was released openly and the more that the papers really were released openly, the more likely it was for people to help be able to analyze that type of data. And so I've sort of been involved in this even since I was an undergraduate, some various types of openness. So open access to publications is really critically important for the spread of scientific knowledge, both, so I worked at this genome center, it was leading the genomic revolution, but we only had two people, and we had no subscriptions to journals, basically. And it was crazy, we're doing this cutting edge genomics work, and I couldn't get access to the scientific literature about like anthrax, and it just was incredibly frustrating. That's when I started getting involved in the open access to the scientific literature movement and it's just really, really important for scientists to not have constraints in what they are able to read and what they're able to share with other people and students as well, to have that scientific knowledge and literature available. And I think the same is generally true for scientific data and other scientific tools. Again, there are, the question was about like companies and the balance with companies. Most of the companies that I've interacted with also want access to all of the scientific literature. They don't necessarily wanna share their own private data, but they certainly want to have access, and I think they should have access to publications and data that are generated by public funds. So I don't think we should tell every company that they need to release their private data about particular studies, they're not gonna have incentive to invest in particular types of work in many cases. The NIH puts billions, $20 billion a year in biomedical research and the NSF puts $2 billion a year or something like that, and the goal of that scientific knowledge and that material, that data, the publication should be freely and openly available and it certainly in the case of pandemic, it's really kind of sad and interesting what happens, what happened with the flu outbreak, I blogged about that 10 years ago complaining and happening right now is closed scientific journals that do not make their scientific literature freely available. When bad pandemics happen, all of a sudden they announce, "We're gonna make our publications freely available "about coronaviruses for the next three months. "Oh, aren't we great, we should get a big pat on our head "and everybody should blow trumpets for us "because we're making this material freely available "for a few months." And my response to that is, "We would have been better able to respond "to the coronavirus outbreak if people have "had access to that literature before the pandemic, "not after it's happened." So it's great that they're making it available, and it's true that it is very useful to have it all available during a pandemic. It's also useful to have it available at other times.
Q: Just sort of building off that, I mean, when possibilities that there's a lot of research that goes on in companies, that turns out to be negative results that never get published and never get put out there and would be really helpful to have for the fields to progress knowing what companies have already started. I mean, I think the original SARS had a vaccine attempt that kind of went nowhere, I'm not really sure if there was really good studies on that or not. So, I mean, that is the other side but there's this empty halt, this black hole where probably good experiments are going and companies and not being reported for one reason or another, at least that's my view. So I'm curious what you think about that.
No, I mean, that's that's actually true. I've been fogged in helping some people create these. There are a few journals that are publishing negative results. There are some journals, many journals like this one called PLOS One and there's one called PeerJ, which the principles behind the journals are that they will publish anything that is scientifically sound, and have to have a new result that could, in fact, have a negative result. It doesn't have to have a positive result. They will publish anything that is scientifically sound. And it is critically important that we do that. And part of the complication with companies is there's very little incentive for them to publish those negative results to necessarily help them in many ways. So what we need to do is find better ways to incentivize everybody, both academics and people or companies to publish those types of results, either, because they get credit for an added result in some way or because they got some additional funding or something like that. Because if there is a huge bias in a lot of the literature and not publishing that, there are movements for clinical trials to require people to publish the results of clinical trials, whether or not they're negative or positive but that has not spread enough.
Q: Negative results, meaning no correlations were found? Yeah, so there's data that did not support the hypothesis. Oftentimes, it just doesn't get reported because it's not looked at as being a sexy result, for lack of a better word.
Or even there's a similar thing of a negative result, which is a confirmatory result, which is also frequently not published because it's no longer considered novel. Both of those things, not publishing something where you did not, could not refute a particular hypothesis, or did not get enough data to address something or that you just confirmed someone else's work. There's also less incentive to publish that, and we need both of those to be published.
Q: It's related to the spread of COVID-19 in California. And that there's a theory that it's dormant. Why are there not as many cases in Yolo County as elsewhere? maybe you wanted to touch on that.
Well, I'm not sure there was a bunch of news stories that have gone viral, so to speak, in the last day or two about this idea that some people were proposing that in the fall, that people were getting sick and that might have been related to the coronavirus having already spread in California. There's an excellent article in Slate that was published today basically saying that that's a pile of garbage, that there is just absolutely no evidence that the coronavirus showed up in California in the fall prior to, in essence, a little bit later in January or February. So, it's certainly the case, know a lot of details about that fraction of the population that has been infected in California with the spread that's been happening now. We don't know a lot about, in essence, how many people have few symptoms or no symptoms at all and the age distribution of those people with no symptoms, and that's because we haven't done a lot of testing within anywhere in the country, anywhere in the world, really except for a few places. So we don't know a lot about the baseline. Unfortunately, right now in California and other places is people who are sick enough that they either wanna get a test or go into the hospital and get tested there. And so we just don't know a lot about ground full of infection in a lot of places. There are finally some surveys that are being done that are addressed that and start to get information about the background level of infection. And also, you can use antibody tests to look at whether or not people have been infected in the past. And those tests will be also being ramped up in the next couple of weeks.
Q: I'm not entirely sure where it came from, the idea that a virus could go dormant. So I mean, maybe we could touch on that idea that as the virus spreads, is there a possibility that it burns out somehow? I mean, obviously, there's an immune response that begins to take over in the population and how adaptive the virus is.
So this goes back to the question from Santiago before about what can we learn about other emerging infectious diseases, which is that in some cases, in prior cases of emerging infectious diseases, a new pathogen will emerge either because it jumped from animals and humans or underwent some type of weird mutation process like happens with recombination flu, and then it's very virulent, very dangerous, infects a lot of people. And it can undergo evolution and a process that is sort of generally called attenuation where actually, some of them can become less dangerous over time. And that happens for a variety of reasons. For some types of pathogens, it's actually beneficial for them to not make people as sick as possible, because for some types of infection it allows them to actually spread better. If you make someone really really sick, like with Ebola, Ebola virus attack has a hard time spreading in part because it makes people so incredibly sick that it's obvious that they're sick and people wanna stay away from them, also don't travel and they don't do a lot of things. If a virus was, sort of related question, if it was much less dangerous, it might be able to better spread and that can be a way that natural selection can make certain types of pathogens less damaging over time and that has been seen with a variety of other infectious diseases in plants, animals, and in humans. There's no evidence unfortunately that that is going on with the new coronavirus. It's certainly, there's no evidence that virulence has gone down in the course of this infection, it still could. But there's no evidence at this time it has. Beyond that can happen, of course, as you were saying is, if enough people in the population become immune, the spread that can, the rate of spread can go down, because then one person who is sick bumps into another person, if that person is immune, it doesn't spread to that person or spread to another person from that person. So that's sort of the general concept of herd immunity. Almost certainly, we have not reached the level of immunity required or big, significant amounts of herd immunity in the United States.
Q: Somebody read that in Iceland, they're doing widespread testing with a large population. The infected people turn out to be asymptomatic. Could that'd be true here? And obviously, what would that mean for the future of the pandemic?
In Iceland, about 10% of the population there. That's what I read in a news story yesterday maybe and they're not completely randomly testing, but in many cases, they're basically randomly testing people. You can get good ideas of what the population frequency is, not biased by who's sick or who's not sick. And so it is true that a huge number of people that are testing positive there either have very few symptoms or are completely asymptomatic and almost certainly the same thing is happening here. There's no reason to think that it would work differently in Iceland and then this is also shown in some testing that was done in China and a few other places. So it's very, very likely that a large number of the people that are infected are either asymptomatic or have very subtle symptoms that they're attributing to allergies or minor other annoyances. And I don't know if people know but the original reports on this virus and infections focused on the initial symptoms being fever and cough-related and then progressing into lung infections and pneumonia and possibly other issues. There are huge diversity of phenotypes that have been associated with this virus that people have been recording and some people have no fever, some people have no cough, but they have digestive issues or they have a loss of smell. People may have heard of that thing occurring or there are a variety of other symptoms that this virus appears to be causing and it's just unclear why some people get some symptoms and why people get others and why some people may get no symptoms. But that's one of the really hard things, right, is if we wanna shut down the transmission, if a large number of people are asymptomatic, but infected, the only solution to figure this out then is testing, is ramping up lots of testing to be able to control the spread.
Q: So advocating for sequencing large populations of microbes, in this case, viruses could be quite powerful. But on the host side, there's clearly genetic variation within the population, whether it's epigenetic variation or genetic variation, there's clearly something that's changing how people react to the virus. So maybe you could comment a little bit on those questions and that idea?
Yeah, I mean, there's not a lot of data yet reported on the genetic side of susceptibility, but you can basically imagine almost any trait of the host and that would include genetics, health status, in particular, the immune status, certainly, the functioning of immune system and prior exposures to particular pathogens or relatives of the coronavirus. And then a variety of other features clearly are related to susceptibility. The one that's gotten the most coverage is age. Age is clearly an incredibly important factor in determining the severity of the infection in people and there are a lot of hypotheses about that. Most of them relate to the functioning of immune system, but I don't think people actually know what's going on there. I am type I diabetic. Diabetics are reported to be at very high risk to severity of infections. It's not clear if that's type II and type I diabetes or possibly just type II, I don't know. But I'm in a very high risk category. It's why I'm sitting in my yard and not going and interacting with any other people other than my family, and not letting my family go interact with anybody. And so it's not clear why diabetes is a particular risk factor. There are models that explain it. And so you can add on top of all of those things, genetics certainly could contribute to risks, either you have some polymorphism in the receptor for the virus. So the virus binds to a particular type of protein and cells of these are some paths for infection. It does all sorts of other things, and any genomic, genetic variation and any of those factors could lead to differential susceptibility. But as of yet, there haven't been a lot of papers on that side of the story. I'm sure they are coming any day.
Q: There was some cases apparently in China, Japan, South Korea who were diagnosed with COVID-19, recovered, and then readmitted after testing positive for the virus. I guess, the general question is about getting past the viral, the disease and being exposed again, or coming or somehow coming down with it again. Do we know anything?
No, I've read many of the papers behind that and a ton of the news stories. This is an incredibly important question. One of the most important questions about what's going on is after you've been infected and recovered or potentially recovered, are you immune? Are you immune to subsequent infection? And we actually don't know the answer to this question, which is a little bit scary. If people are not immune after being infected, they can get reinfected. And right now, we don't yet know if people are going to develop a short-term or a long-term immunity based upon infection. However, those reports about reinfection of possible people in all cases that have been published so far, these are people that were very recently infected. I mean, it's a new outbreak, so most people are very recently infected, and they have been reported as cleared, that is, they've tested negative, maybe twice. But the way the tests work, there are lots and lots of false negatives for the tests upwards of 60% to 70% of some of the tests have reported incorrect, false negative results, depending on which test is being used. So most of the people that I have seen who are virologists or epidemiologists or infectious disease people think that the people that are coming back into the hospital, saying they are now infected again after being cleared were never actually cleared of the infection. That's what most people think. That is, they tested negative, but it was a false negative or they just had a low amount of virus in them, but they hadn't yet cleared the virus from their systems. So there's no evidence yet that anybody after being infected and completely cleared can get reinfected quickly. That has not been shown yet.
Q: Students wanted to know what tools were being used to sequence genomes of the viruses, RNA viruses, how do they work? And maybe maybe shed a little light. I think we're gonna get to this in the course on some of the tools that we use to try to detect the viruses, but there's a whole lot of things going on right now. And so maybe a quick discussion of that, from your perspective would be helpful.
If you wanna detect the virus in a person, there are basically three ways that people try to do this are three main ways. One is with an antibody assay, where you basically take a sample and then you expose the sample to antibodies that are known to recognize this particular virus and you can show whether, that's sort of like how home pregnancy tests work and a variety of other things. You can show that there's a positive hit to binding to that particular antibody. Another way is you can actually grow the virus in tissue culture in the laboratory, so you can actually see if you can infect cells in the laboratory and quantify or characterize whether or not the virus is present in samples. But most of the tests that people are using for detecting the virus involve looking at the presence of the RNA from the genome of the virus. And the way this works is basically you take your swab or your sample, you extract, you run a chemical reaction that extracts nucleic acids, RNA out of the sample, and then you run some molecular detection method like the polymerase chain reaction to see whether or not you can detect the presence of that specific RNA in the sample. If you wanna go further and sequence the genome of the organism, not just detect whether or not it's present, it's very similar. You just take the products have that PCR reaction and then you run DNA sequencing reactions from those products. So there are probably 50 different approaches that people have developed to doing this already. Sadly, unfortunately, the first approach in the United States that was recommended by the CDC did not actually work really well. That created a bit of a time lag in some of these but there are now, UC Davis developed a method that was just booted up about a week and a half ago. They have a set of equipment that they use. They can do about, I think, 200 tests a day now. UCSF developed their own protocol. They can do about 1,000 tests a day. Berkeley developed a protocol. Stanford, I mean, like, everybody's developing protocols, and again, going back to the Open Science Movement. Fortunately, everybody is also sharing the details of all of their protocols so that people can boot up new testing facilities really quickly.
Q: Would it be reasonable to have the population of California tested randomly to really get at this question of how good our tests are, and whether we can detect that asymptomatic population, it would inform our modeling a lot better? Is that something that we have the capacity to do and you would advocate for if we do have that capacity?
Oh, I would certainly advocate for it. It depends on which type of tests you wanna do. I think, very soon this test for whether or not you may have the antibodies in your blood are gonna come online. That is, if you have been exposed to the virus and produced antibodies in the past, those tests are probably gonna come online soon and they're certainly the potential to do hundreds of thousands of those tests because it works in the same way as a basically any other blood test. And there are thousands of labs that can do blood tests. They don't have to do anything different. The DNA detection assays, and here's the really weird and interesting thing, I guess, about these times, it's the PCR part of the process is not in fact rate limiting. Unfortunately, some of the things that are waiting, I'm gonna include swabs. The swabs that are used for nasal pharyngeal sampling are in limited supply right now. And so on this conference call that we had earlier today with the people in the hospital and all across campus that are working on this, they said that people in, I think, biomedical engineering department, 3D printing swabs to now use for some of the nasopharyngeal sampling because all across the country, people are running out of the swab do this test. So it'll still take a few maybe weeks to solve some of these supply problems and get everything ramped up enough to do a lot more testing. Fortunately, the governor of California announced a task force focused explicitly on ramping up, massively ramping up testing within California. This was just like four or five days ago, they announced this task force. And that's one of the things they're gonna be working on is like, what are the rate limiting stuff? So another one turns out to be the chemicals used to extract RNA from samples. The kits that people were using for this, you would buy basically a box that came with 100 little tubes that were pre-selected with this RNA extraction buffer and those, everybody's run out of all of those. And so, people are scrambling trying to find the reagents to do the RNA extraction. I'm on the Scientific Advisory Board of a company that makes some of these reagents. And amazingly, this was like two weeks ago when this was even a bigger problem, I was cold called by someone from another state running the entire Public Health Division in the other state who was desperate to get RNA extraction reagents and wanted to know if I could connect him to the people at this company and sold RNA extraction reagent. So it's just amazing how, I mean, sadly, how unprepared we were in being able to deal with a pandemic of this proportion.
Q: There's a bunch of companies out there that are trying a variety of approaches. And the student was asking how it affects the estimates until there will be a vaccine available for public. I know, I saw things in my newsfeed this morning about it and didn't have time to read it, so I know this is a hot topic.
Yeah, I mean, again, I've helped with this a long time ago with bacteria-related vaccines. I don't really do a lot on it right now. So I don't know a lot about, they're just, it's amazing. I mean, there are probably, I would say at least 50 candidate vaccines being tested throughout the world by different companies or different academic institutions, probably more than that. And you know, it's the normal path for developing a vaccine for a new infectious disease is years. Like three, four, five years. The estimate that Tony Fauci and other people made saying, we might be able to do this in 18 months, is considered optimistic by many of the people who have worked on vaccines previously. But if you actually look at what they're trying to solve there, there are some reasons to hope that things could be developed a little quicker than the three-year or even then the 18-month development process. Because some of the steps that have been limiting in the past, they can get around some of those issues. For example, they're not, I mean, it's a little bit scary, but they're not doing as much animal testing. They're just jumping right into human testing for some of the vaccines. The funding, it can take a while to develop the funding and develop the cohorts of people that you would use for developing a vaccine for a new infectious disease. There's an enormous disincentive for people to collaborate on this right now because of the degree of severity of the pandemic. So like Bill Gates announced, he's putting, I forget what it is, a billion dollars into fund the distribution plans for vaccines, 'cause one of the problems is not just developing and testing a vaccine. But once you actually show that a vaccine works, you have to make it for, if it's the United States, you have to make 300 million doses of it. It's just unbelievable how complicated that is to do it on a scale. And if we wanna make it for the globe, that's billions of doses.
Q: I've seen articles flying by my news feed, vilifying companies about the bottom line and whether they're truly being helpful in making vaccines. Do you have any sense about this? It's very hard to read the news and know what to believe or not to believe.
Oh, I mean, I think, just like in any other part of human endeavors, there are diversity of responses to these types of things. So, I mean, I saw some of the news stories about one company in particular that was making ventilators and was reluctant to share the specs for their particular ventilator that they were developing. And I think they had some contract that they eventually signed with GM or Ford or some other organization to make their ventilator, and I'm sure it's a complicated calculation that some companies are making to try and decide like, how open do I wanna be versus how do I wanna try and actually get a contract to do this from some other entity and they wanna stay in business? So, I mean, I understand that it's a balance that people have to make. I think that I'm generally impressed with the amount of sharing that is happening on companies related to ventilators and vaccines and data and other things. So I think they've been, overall, way better than in normal times. And they are responding mostly very positively and very sort of selflessly to what's going on to try and help out. So overall, I think it's been pretty good, but there certainly are some companies that are not helping as much as they could.
Q: So speaking of understanding this and getting the right information, one of the questions from students, what sources do we, as professors, and I'll let you answer it first, anyway, find the most accurate to stay up to date with whether evolving information on the virus and the biology or other sources, I guess?
Yeah, it's so hard, I mean, even, so it's a little bit funny and weird for me, in a sense, because my background is basically genomic epidemiology of microbes and evolution and phylogenetic of microbes and high characterization of pathogens and all these things that are right at the center of everything that's going on right now. I have a really hard time, the amount of information is astonishing, the rapidity at which stuff has been astonishing. So for example, just one little niche in that area, the publication of what are called Preprints. So Preprints are a relatively new thing in biology. They've been around for a long time in physics and math, but in biology, they really took out 10 years ago with the creation of a place called bioRxiv. And what they are, are people releasing, in essence, a PDF of their scientific publication prior to it being peer-reviewed, prior to it being published in an official published in a journal and they're releasing it to help spread scientific knowledge or to stake out territory or whatever it is, they're releasing it early and really rapidly and this is been amazing in the case of this pandemic. There are dozens to hundreds of new COVID-19 related papers posted on bioRxiv every week. And then, equivalent version called MedArch now, and there's archive, and there's just been tons and tons of releases. And the problem is that there are way too many for even the experts to screen right now. And so I've been reading a bunch of them in my particular area that we're working on related to the virus, which is the viability of the virus in the built environment on surfaces and the transmission dynamics of the virus. We're actually going to do work on this. And there are like, a day on this topic, and some of them are like 70 pages long, seven tables and it's just, I can't keep up. And when I read some of them, I'm like, "Oh, this is just completely bogus," or "This seems realistic," just it's really hard. So I think what you need to do is find trusted sources that you really like or trust and keep reading them. So I have a list of people that I follow on Twitter that I think are really good about this particular area and I read things in particular newspapers or journals that I think are really good.
Q: I do think it's important for students to know the peer review process and to be able to understand bioRxiv is a great place to go see papers, but you have to understand they have not yet been peer-reviewed, doesn't mean they're wrong, it just means you have to take them with a grain of salt.
Peer review is not saying whether or not a publication is correct, it is basically two or three people that read a publication and try to do some type of screening of the publication. Lots of things get through peer review that are completely bogus. Not as some things get caught by peer review and peer review can add quality to scientific literature, but it does not mean that a publication is correct. So the people that say, "Oh, after peer review, we can trust something." That is not true. You still need to be skeptical of anything even if it's gone through peer review, but certainly, things that have not gone through peer review, you should be more skeptical of.
Q: I'm gonna take you back a little bit to phylogeny. So there were a couple of questions early on about viruses being alive or not? Where would they fall in the tree of life? I think this came from one of your BIS 2C students.
Well, if that was, she should know that. So there are three main theories about where viruses come from, and they're probably all true. They just don't all apply to all different types of virus. The diversity of viruses is just enormous. There are so many different kinds of viruses. And it's unclear and it's almost certainly that the viruses are not all related to each other. So the models are basically, the models are that viruses represent highly reduced cellular organisms. So it started as a cellular organism and threw away a lot of its components, including the cell itself and it's left with just some machinery basically to spread between organisms. There's a neural model that is viruses are, in essence, an escaped component of a cell, like a transposable element that is a piece of DNA that moves between different parts of the chromosome. All you need to do is add a little protein package on the outside of that transposable element, and in essence, it can function in ways like a virus. And that certainly seems to be the case for some things like HIV, which are known as retroelements. They're very, very similar, in many ways, to what are known as retrotransposons. And then the third theory is that viruses are relics from the origin of life, and that is they represent another branch in the tree of life that is leftover from early, early evolution. When we used to think that there were three main branches in the tree of life, the so-called three domain tree of life. Some people said that viruses represented the fourth domain, the fourth branch in the tree of life. In fact, the main person behind theory was Didier Raoult, the hydroxychloroquine researcher. These and I know him as we published a paper suggesting that there were some weird DNA sequences out in the environment that might represent a fourth possible branch in the tree of life. And he invited me to come give a seminar in Marseille, which I did about five years ago, and he was my host there and we talked about weird viruses that might represent ancient relics from an old time. And all of those theories are probably true. Some viruses are probably leftover from billions of years ago, maybe representing a novel branch in the tree of life. Some viruses probably represent escaped portions of individual cells and some viruses probably represent highly reduced versions of cellular organisms.
Q: Where are you seeing this pandemic in terms of scientists interacting with the general public and maybe regaining reputation or moving into a new direction where experts in very specific areas of research can actually engage with the general public and educate the general public and what were you seeing also managing misinformation?
I think, it's kind of both things. I mean, like, scientists have been doing a really good job, I think in the last few months in trying to engage both other scientists and the public in discussions of things related to the pandemic. I mean, the number one area that people really did a lot on was this whole idea of flattening the curve, of trying to reduce physical and social distances in order to reduce the rate of spread of the virus. And so many scientists that I know spent a lot of time in their local communities trying to convince people that this was important to do and I think they did a really good job of it. I think it's hit or miss whether or not they're supported for doing this and are getting attacked for doing this in some cases. So there's still an anti science component to the response that a lot of people are experiencing. But I think, in general, there's been a increase in the interest in actually interacting with and hearing from experts, people with actual expertise on the topic of epidemiology and infectious diseases. And so for example, a few of my friends who probably came on to Twitter kicking and screaming four or five years ago, now have follower counts of like 300,000 because they've been heavily involved in commenting about the epidemiology of the coronavirus and interacting with the public in a way that people really wanna hear what they have to say. So I think I've been impressed with how many of the people who are good scientists working on this topic have embraced the need and actually learned how to do a good job at communicating right now.
Q: Emily’s asking that a lot of places in California now require wearing masks when you go out and the question is, do masks actually help prevent someone from getting infected? Because I have heard that the virus is so small that it can pass through. And this may be a question that you may be able to answer given your expertise in this.
I've been following this quite a bit in the last couple of weeks. There have been maybe 10 scientific studies published in the last two weeks about this topic. They're not all consistent with each other, so there are some ambiguities here on exactly what's recommended and what's useful, but I think it's pretty clear from the, if you look at the full diversity of all the publications and reports that certain types of masks like surgical masks can certainly reduce the probability that someone who is infected will spread the infection to other people around them. It doesn't eliminate the possibility but it reduces the possibility. The homemade cloth masks that a lot of people are using or other types of cloth masks are not as good at doing that as like surgical masks are, but they still appear to be likely to help reduce the transmission. Whether or not the masks protect individuals from getting active themselves depends a lot on how exactly you use the mask and which type of mask, you know. If you go out into a grocery store and you're wearing a surgical mask, it probably can protect you from getting some types of exposure. But if you then, when you get home, if you've been touched the outside of the surgical mask when you're taking it off, or you put the surgical mask in some environment that you're gonna bump into later, that doesn't necessarily help you. I saw a good guide to dig around for, a good guide on how to use masks and how to wear them, but they certainly seem to be likely to help reduce exposure to droplets. So here's the complicated part of the story. So when the infection, when the pandemic first started, everybody said that the main route of transmission was via droplets from an infected person that got onto a non-infected person or got onto a surface that a non-infected person then touched with their hands and brought in. That's why everybody said wash your hands all the time and that's why people said you need to stay about six feet away from people because the droplets like if I'm coughing or sneezing, the droplets generally will eventually plummet due to gravity and if you're far enough away from someone, you won't get exposed to those droplets. If the person's hands are infected and you touch their hands, you have to wash your hands. If you touch something someone sneezed on, you have to wash your hands. So that's why they recommended washing your hands and staying a reasonable distance away from people. However, there is growing evidence that the virus is also airborne. That is that in addition to droplets, it gets aerosolized and then can spread a little bit, at least, somewhat in the air further than six feet, that the droplets would spread. And the problem with this is that the masks also, except for the N95 masks if you wear them correctly, all the other masks don't protect you against the aerosolized version of the virus very well. And it's just unclear right now whether or not the aerosolization route is an important route for transmission of the virus. But I think it's worth worrying about that for now and trying to be really careful about it, it's not just staying six feet away from people. You don't wanna be in an enclosed environment with a lot of other people, even if you're six feet away from them for a long period of time. So if you're in a small room like a bedroom and you're standing on the opposite side of the bedroom, six feet away from someone and your droplets are getting on to you, there is probably gonna be stuff in the air in that bedroom that you might get exposed to. If you're outside, it's less of a problem, makes sense. So it's just unclear. The publications that have come out in the last week, some suggest that aerosolization is actually a really big worry, and some suggests that it's not a big worry, and we just don't know right now.
Q: So quick follow up on that, what about the possibility of doing environmental sampling like you've done in the past? Like to just, to know what areas, like can you go to a supermarket and then swab every surface that people touch, and then detect the virus?
Interesting that you asked that. We have put in four grant proposals on this in the last week and we spent, I spent, people in my lab spent a few hours yesterday talking about doing environmental sampling. We're actually gonna start our first environmental sampling for COVID probably in the next day or two in collaboration with the UC Davis Hospital. And we're trying to develop the techniques that would be required to do this, so that we can then do surveys like you're saying, we would like to be able to survey other built environments for the abundance of the virus in those environments and in air samples in the environment, and we and many other people are trying to ramp up the ability to do that type of environmental testing. The challenge right now is really we don't wanna, we don't wanna take away from the clinical testing that's needed. And unfortunately, the environmental testing right now requires the same exact reagents that the clinical testing requires. So the main thing we're trying to figure out how to do is, can we do the environmental testing with other types of reagents that don't take away from any of the clinical testing that's needed? That's what we're working on.
Q: And there's a window of time to it. You can't wait too long before you can't do the environmental testing. So there's a finite window, which you can do this, I would guess.
I think, it depends on what environment that we're interested in doing that in, but I mean, I think certainly, we're thinking for the next six months, the next year, in addition to wanting to test a lot of people, we're also gonna wanna know how effective are our cleaning practices and where are the hots transmission that are not due to someone's sneezing on you.