Hello, and thank you for listening to the MicroBinFeed podcast. Here, we will be
discussing topics in microbial bioinformatics. We hope that we can give you some
insights, tips, and tricks along the way. There is so much information we all
know from working in the field, but nobody writes it down. There is no manual,
and it's assumed you'll pick it up. We hope to fill in a few of these gaps. My
co-hosts are Dr. Nabil Ali Khan and Dr. Andrew Page. I am Dr. Lee Katz. Both
Andrew and Nabil work in the Quadram Institute in Norwich, UK, where they work
on microbes in food and the impact on human health. I work at Centers for
Disease Control and Prevention and am an adjunct member at the University of
Georgia in the US. Hello and welcome to the MicroBinFeed podcast. Today we're
doing a rapid roundup of the latest information you need to know about SARS-
CoV-2. Today in the booth, we're joined by a special guest, Leo Martens, who's
head of phylogenomics at Quadram Institute and our Arborist-in-Residence. And
then there's always Lee, Andrew, and myself. And guess what? Everyone's working
on SARS-CoV-2 now. And so that's what we're talking about today. We'll be
focusing on some of the latest and greatest resources available for SARS-CoV-2
analysis. So let's kick off with some news. Andrew, you ran into something
interesting. Yeah, so I thought it'd be a very quiet week because a week ago,
right, all the focus was on other things than genomics. I thought, great, you
know, we can actually get some work done in peace and quiet and everything will
be fine. You know, there's the vaccine wars. Only a week ago, we're kicking off.
Late on a Friday night, the European Union drunkenly kind of wandered into this
minefield of Northern Ireland politics. And my God, if any Irish person had been
anywhere near that, they would have told them that it was a very bad idea. But
of course, it didn't happen. Then, you know, there's a huge hoopla over all of
that. And you know, we had vaccine wars kicking off. And great. But
unfortunately, there has been a huge amount of progress now in genomics. And
everyone wants genomics, SARS-CoV-2 genomes for all their work. And everything
is about variants. And my God, it's been a very busy week. So let's talk about
just all the new resources and stuff that's come out just in the past week or
week or two, because it's been very, very busy. Yeah, it definitely seems like
there's been a massive switch over to trying to capture these variants and track
them as quickly as possible. I mean, I think we can kick off with the
covariance.org page, which for me is super interesting. I mean, Emma Hodcroft is
leading the charge on this one, but there's a lot of people who behind like the
sort of Nextstrain group and people from JSAID and so on are contributing to
this. This is covariance.org. And we'll have a link to that in the show notes.
But this is a very pretty website that just catalogs all of the different
variants and all these different mutations. And you can go to each of the major
ones and have a look at which mutations are there, what these mutations may be
conferring, like what the previous literature said about these mutations. So
this one might alter recognition of antibodies, or this one might change some
structure of the spike protein, or this might do something else. And so they've
got a breakdown of all of the variants and then what all of those are possibly
doing, which is excellent, because now I don't have to look up every time my mom
asks me like, is this one more dangerous or not? I can just go here and just let
up. This one does this. Yeah, this one is more effective or this one is more
deadly or something like that. So it's an excellent thing to have a look at. I
guess there's so many different variants out there, you know, thousands upon
thousands. We don't know which ones are going to be a problem usually until long
after they've emerged and have spread everywhere. Websites like this do really
help because it allows you to see the signals earlier. Yeah, and it's definitely
once one of these variants are flagged, they have a breakdown of these variants
as they, you know, its distribution in per country or the distribution of the
different variants in a country, which is all backed by the data that's coming
into GISAID. And it's just graphing that for you so you can quickly jump to see
how quickly it's spreading once there is one of interest on the website. And
it's good to see that lots of countries are now contributing more so than they
had in the past. I don't know where they're hiding data away in the basement or
if they just weren't doing sequencing, but it's good to see that there is more
distribution around the world of sequencing because, you know, the UK is
dominated a lot. And if you look at some of these signals, you might think, oh,
God, the UK is like the center of death in the world, you know, and plague for
COVID. But actually, it's just they're the ones, you know, doing the sequencing.
So a lot of stuff is biased towards them. Maybe there's also the case that some
countries or some groups were worried about, you know, research parasites. So if
you are the first to add data to the database, and then you feel that people are
going to use the resource without proper acknowledgement. But now that you have
so much data, I think you don't, right, that the priorities change it. You're
not so worried anymore about. Yeah, it still is a big problem. People complain
about it. People complain about it as well all the time. But well, you know,
it's that trade off for being really open and getting stuff out there quickly
and, you know, maybe getting scooped. It's an unfortunate bit, but public health
is more important than retrospective academic research and credit. So another
really cool report on in the same kind of vein as the covariance stuff is only a
tool has a fantastic daily report on COV lineages. And so she's gone through the
major variants of concern. So that would be I'm not going to use the place names
because that's super controversial. But you know, you have the B117, which is
associated with United Kingdom originally. And then you have B1.351, which is
associated originally with South Africa. And then you have P1, which is
associated with northern Brazil. So she's been tracking things like, say,
passenger numbers out of countries, you know, during the time when those
lineages are expanding and, you know, what countries have been identified in. So
you can kind of predict where the holes are in the world and which countries are
probably most affected. So you know, for a while there, there's one particular
country, I think Poland, which had huge numbers of passengers from the UK, but
zero recorded cases of the UK variant. I was like, that's a bit of a problem
there. The surveillance there hasn't picked that up when it should be. And you
can see that with some data reports, it does give kind of an early warning and
an indication of the spread. And it's quite interesting, you know, some of the
places where these variants have been found that the P1 has turned up in the
Faroe Islands. Now, the Faroe Islands, if you don't know, are these teeny tiny
little islands in the northern Atlantic, just kind of, I suppose, north of
Scotland, you know, that's how far away they are. And they're a tiny, tiny
population, the population of a town, more or less. And somehow they have a
variant popping up there. But you know, they've obviously got a strong
surveillance system there that picked it up. Oh, just a stupid thing. I just
learned that my geography sucks. And wow, Faroe Islands. But you actually
pronounce it correctly. It's interesting because sometimes there's football
matches, international football matches, you know, between local countries,
particularly in the run up to the World Cup. Now, Lee, you probably don't
understand any of this as an American, but you know, football is in soccer. So,
yeah, there's occasionally football matches and, you know, you'll have England
playing like Faroe Islands. And it's like a population of, what, 60, 70 million
versus a population of probably 50,000, you know. And it makes for a very, very
uneven game. But, you know, they're a country, so they get to be represented in
that way. There's a joke from my university, Emory University, that our football
team is undefeated, and that's American football. And there is no football team.
That's as much as I understand from football or American football or soccer or
whatever you want to call it. It's like sequencing, you know, no sequencing,
sure, you've got no problems. So, yeah, I really like the cobbling stuff from
Anya. I mean, it's her and the rest of the guys up in Edinburgh putting this
together. It's fantastic stuff that they've got. It's not just the spread of the
lineages, but they've actually merged it with sort of anecdotal reports of it
moving to, you know, people, reputable people mentioning this stuff on Twitter
or local newspapers or so on. And they've even pulled in what the travel, sort
of the number of passengers traveling from different countries, just because
that data is available, who travels to where. Now, even if you don't know that a
particular variant is in that country, it's quite likely that it's going to jump
between countries that have a lot of passengers flying back to the ports. We're
starting to get into this minority report thing where we can pre-cog, like
predict possibly where it's going to spread next, because we're aggregating all
of this different information. And it's presented really nicely, a lot of these
thoughts of the kind of things that you can just copy paste straight to the
newspaper of what's going on. And I think it was interesting to see for the
B1351, was it that?  one that wasn't reported in the US and then it was reported
like I first saw it on this website because this is just running automatically
in the background pulling in new genomes and just generating the figures. Is
that the South Carolina one? Yeah I think it was the South Carolina one yeah
that just flagged up on the website that was the first place I saw it. So
following on from that there have been some recent updates to MicroReact. Now
MicroReact is again another visualization program that is showing you the tree
and the timeline and the spatial positioning of all the different samples and
what's really interesting with it now is they've now got a sort of when you go
to the SARS-CoV-2 section there's simple links that jump out of all of the SARS-
CoV-2 genomes out there it immediately jumps to these new lineages of interest
new variants of interest and then you can just see exactly how those are
spreading on their own. That was one of the new updates so that's from Anthony
Underwood and the rest of the guys at CGPS and the other thing that they've
changed recently is now they show the frequency of a particular lineage of the
total number of samples there rather than just the absolute counts because it's
sort of like you wound up with this issue where you would see lineage is sort of
dropping but it's not actually dropping it's just you don't have the same
absolute count of sample numbers over time and so it looked like the numbers
were dropping but it wasn't and it needed to be kept as a sort of proportion of
the total so that's also been added now as a as an option so both of those make
it like really easy to jump straight to particular variants and see exactly
what's going on really quickly. And of course they get it like a live data feed
from COG UK so their data is usually more up-to-date than what you can get from
GISAID. Now they're not accessing secret data or anything it's that's the data
that's on the COG website it's just they're kind of pulling it in more regularly
and that's from David Aronson's group and it's fantastic really they you can
actually use this yourself as well like with microreact you can just kind of
drag and drop in your your trees and stuff and metadata and it'll give you
really pretty pictures they've made that you know super easy and it just works
really really well for coronavirus and I'm really impressed actually how it's
scaled now to hundreds upon hundreds of thousands of genomes we're up to what
nearly half a million now like it's insane like it's it's a lot. I think there's
some pretty aggressive pruning in the tree or something just to make the
rendering catch up on it but it's yeah it's it's super intense if you open up
the SARS-CoV-2 section and you've got the map you've got all of the pie charts
of the different variants circles on all over the map you've got this sort of
admixture thing at the bottom of all of the of the different variants as they're
coming and going of the total samples the tree I mean this is the I don't
remember in Contagion in the film if they had a nice little dashboard like that
with all of these like the the samples like kicking over but it is very much
like that. I think they only had like a floating piece of DNA double helix or
something. Maybe they had a protein structure like that was just constantly
spinning I can't remember. What you're saying is that microreact is great for
doing these background pictures of scientists right looking at looking busy in
front of a screen. Well yeah I mean microreact has been featured with Ted's
estate now I think. The New Zealand premiere? Yeah yeah New Zealand PMs had a
look. I think the UK prime minister's also stood in front of it but don't quote
me on that. I think he also has there's a few pictures of that so you know it
looks nice but I think the most important thing is that they've changed aspects
of it very quickly to adapt to this new life cycle now where we're pulling out
individual variants and now you can you can not only just have pretty pictures
but you can actually find out what you want to know very quickly. So another
thing that's arisen this week is different variants of concern or lineages of
concern getting different mutations from other lineages of concern or them
arising independently and it's now emerged that some people have made up special
names for some of these mutations like N501Y is now Nelly as in Nelly the
elephant. So that's the one that has that is present in the UK in the South
African and Brazilian lineages in fact all of the variants of concern and then
you also have EEC which is E484K which is another very interesting one and
actually so that is in the B1.351 originally discovered in South Africa and also
in P1 originally and P2 from Brazil and that I think is associated with some
vaccines that may be working as well as they should or they hope and that has
now been found in the UK in our own homegrown UK variant B117 which is quite
interesting because these mutations are arising independently and unfortunately
you're probably going to make things slightly worse. So I've heard EEC a couple
of times and you explained it but also it still went over my head can you say it
one more time how EEC got its name? EEC is E484K and when you kind of look at a
written down you think oh okay yeah that that is EEC. I don't know if it's a
thing in America but in the UK people have personalized number plates and they
have to have kind of letters and numbers in particular combinations so people
kind of make words out of these things you know. That's the same thing as with
Doug Douglas and Dan? Yes. E614N. My god now we need another list of all of the
the colloquial names for these variants. Yeah so there is a few really good
useful resources out now. One of them is the Clam Arctic Workshop online
resources so that's like a course that's run for 133 people from like more than
30 countries and it has there's YouTube videos up there's homework you can do
and do assignments just to give you a kind of quick introduction into all that
is how to analyze data you know that your arctic sequencing data. So you've got
really great people like Josh Quick who developed Arctic and you've got Ono Atul
and Verity Hill who've been developing Pangolin and Civet and Llama and all the
other people you know who Nick Loman just just it's like all the rock stars of
the SARS-CoV-2 world have come together and that's that's super handy and I know
Lee you said you have some resources as well? Yeah CDC recently came out with a
COVID-19 genomic epidemiology toolkit. Several people were involved with it.
Greg Armstrong and Nancy Chow, Satavia Morrison, Mike Wigan. I'm gonna stop
reading names off the website but it's a bunch of videos just explaining a nice
primer and it even helps epidemiologists get into it. Yeah some of them are
really nice it's like it's straightforward stuff like how to read a phylogenetic
tree what the gene about the genome applications in Arizona it has here listed
as well. So this looks like super good primer especially for folks who haven't
really dug into genomic epi that much to kind of you know try and get their head
around it. So that's really nice and I think if you put that together as a as a
primer for the workshop which is more practical the client big data workshop
then you're gonna have you'll be able to press out some pretty well-rounded
analysts pretty quickly with this sort of with these sort of resources. I guess
that training and all that hasn't caught up because all the people doing this
you know it changes every week it seems but also all the people doing it are
super busy coping with a million different samples coming through the door that
they have to get out and analyze and whatnot and it's constant firefighting so
you know sitting down for a few weeks to you know fine-tune a training course is
going to be very difficult for for a lot of these experts. So it's great that
the resources are right there but I'm sure things will get refined as as we go
along you know and we'll get more and more resources and it'll be more
coordinated and more coherent. Yeah I can see like a really nice set of
materials that are that are popping up that we can probably synthesize together
to make a nice you know big aggregate course for people to get introduced to it.
So I've seen some PCR methods coming out which just target the or try and type
the variants of concern that are currently concerning and my first thought about
that is that these are probably going to be at a date in a week or two unless
they are going to flip it over just to looking at mutations of concern like
Nelly and Eek rather than actually looking at the lineages themselves but then
you lose the whole genomic epidemiology part of it. I guess you could also go to
an extreme and only sequence part of the spike protein you know with Sanger
sequencing or something like that. What are your thoughts on that Nabil? Yeah I
think it's baby steps with this one. It's I mean we're coming from over
Christmas using the spike gene target failure as the main way of detecting the
B1.1.7. So this and that's a that's a real dirty hack of just I'm just seeing
like oh the spike gene doesn't show up too well with this particular assay so
we're going to go back to go back to  that data. So now we've got people coming
up with protocols. They're putting it up on Protocols.io. We're seeing some
sharing. Maybe this might culminate in Taskforce turning this around quickly, or
maybe some new technologies that aren't so rigid and a bit more flexible. But
this is, this to me is like a great, great first couple of steps that there is
an open community trying to get on top of this. And it does help, I suppose,
countries which don't have the same resources. You know, PCR is a little bit
more straightforward than genome sequencing. Yeah, definitely. I think that's
always a consideration, always a problem that we're going to be using RT-qPCR
for a very long time for just as the primary tracking method in a lot of
situations. Although I suppose Denmark have shown that if you really need to get
a sequencing project off the ground, you can do it very, very quickly. And you
can see Mads Albertsson has got a fantastic lab. It seems to be shoved into a
teeny tiny little room, but they have about 20 or 30 minions running
simultaneously, just wires everywhere, computers everywhere. And they are, you
know, now it's like one of the top countries in the world for generating genomes
per head of population. Actually, they might be the top and they're just doing
an amazing job. And what they've been able to track now is genomics rather than
just PCR. The percentage increase of B1.1.7, obviously Denmark is quite close to
the UK physically, but there's a little bit of water in between, but a lot of
passengers going back and forth. And that I think is a good proxy for what other
European countries are facing because they're up to what, 19% now. And you can
kind of see it go, going up very, very rapidly, you know, 2%, 4% and so on every
week. And tracking that progress with just genome sequencing. So there you go.
Stop whinging and set up your lab like mads. Yeah. It's good to see that they're
getting, I don't know if they're handing it over, if they're getting more
support from SSI on this as well. So that's really going to formalize and, and,
you know, just amplify what, what their efforts are, which is, which is great.
And I think just as a closing final bit of new development is GISAID has
launched a CLI API submission tool. Something I got an email about the other day
previously, if you wanted to put your genomes up on GISAID, you had to make this
Excel sheet or CSV file and have all your sequences in another little multi-
facet file. And then you had to submit it to a webpage and it was good because
you could do batches in one go. So, you know, you do a couple of hundred at a
time, but it wasn't programmatic. It wasn't something that you could just have
as you generated consensus sequences that you could just flick it over to them.
But now they have allowed an API to do exactly that. And that's going to be
excellent for turnaround of, of genomic information. So all of the resources
that we've been talking about, except for the PCR resource is going to be that
much faster at picking up what's going on because the data, our people will be
able to submit the data that much easier. That's all the time we have for today.
And we've been just talking about some of the cutting edge SARS-CoV-2 resources
and yeah, special thanks to Leo for joining us and we'll see you all next time.
Thank you all so much for listening to us at home. If you liked this podcast,
please subscribe and like us on iTunes, Spotify, SoundCloud, or the platform of
your choice. And if you don't like this podcast, please don't do anything. This
podcast was recorded by the Microbial Bioinformatics Group and edited by Nick
Waters. The opinions expressed here are our own and do not necessarily reflect
the views of CDC or the Quadrant Institute.