7 ms·
Genomic epidemiology of novel coronavirus (HCoV-19)
- redcalx 7y agoIran is ominously missing any cases according to this data. I guess you could say Iran is a known unknown.
- throwanem 7y agoIran isn't represented in this dataset. Neither is Africa or New Zealand. I'm not sure what's meant to be ominous about any of this.
- rjtavares 7y agoWe know the virus is widespread in Iran. We don't know if it is widespread in Africa or New Zealand.
- FillardMillmore 7y agoSo widespread in fact that it's even hitting those with high positions in the Iranian government. The most recent example being Mohammad Mir-Mohammadi, an adviser to the Ayatollah. https://www.businessinsider.com/iran-supreme-leader-khamenei-inner-circle-first-death-coronavirus-2020-3 https://www.businessinsider.com/iran-supreme-leader-khamenei...
- throwanem 7y agoNo more so in Africa or in New Zealand [1] than in Finland, which is represented in this dataset. I'm not really seeing the basis for whatever conspiracy theory the original parent comment seems like it wants to point toward, is what I'm getting at here. These are the early stages of what looks to be a pandemic of a disease at least an order of magnitude, possibly several orders of magnitude [2], more lethal than seasonal flu, which is no picnic already. Isn't that enough to be worrying about, all on its own? [1] https://coronavirus.zone https://coronavirus.zone [2] CDC headline numbers for the 2019-2020 flu season, vs. the commonly given 2% lethality guesstimate.
- croissants 7y agoWhat do you mean "no more so than in Finland"? The link at [1] says Iran has 1501 cases and 66 deaths, while Finland has 6 cases and 0 deaths.
- throwanem 7y agoI mean it's no more widespread in Africa or New Zealand, which are not represented in the genomic dataset charted here, than in Finland, which is represented. I've edited the earlier comment to clarify that, since the surrounding context evidently does not suffice on its own to do so for everyone.
- nostromo 7y agoEstimates are dropping down from 2% as we discover lots of cases that were mild or asymptomatic. The New England Journal of Medicine says it may likely be considerably less than 1%: https://www.nejm.org/doi/full/10.1056/NEJMe2002387 https://www.nejm.org/doi/full/10.1056/NEJMe2002387 People should remember the context of a respiratory disease in China. 2/3rds of men smoke, and air pollution is common. Things also spread faster there as it’s densely populated with lower hygiene standards than developed countries (for example: spitting is still common in some areas).
- jfoster 7y agoSuppose you're right about reasons why it spreads fast & lethally in China. How do you explain the fast spread and lethality outside China?
- Enginerrrd 7y agoThe two "fast"'s are relative. Epidemic dynamics are likely to differ in different countries for a bunch of reasons. And, it does appear to be less lethal outside of China. That may be because healthcare systems are not yet overwhelmed, or it may be because of the smoking and pollution exposure in the different demographics. It is probably even both of those things.
- csomar 7y agoAlgeria/Tunisia/Egypt all have declared cases coming from foreign countries.
- icegreentea2 7y agoYes, but this dataset is essentially manually populated. Someone has to isolate coronavirus from a patient, sequence it, and then upload it to GISAID. So the question is why is it ominous that no one in Iran has isolated and uploaded a sequence?
- skrtskrt 7y agoI believe they meant that it's scary that we don't have data in here from a place where we have heard the virus is both spreading quickly and having a high mortality rate. Not ominous as in any way indicative of a conspiracy or otherwise.
- JaggedNZ 7y agoHere in New Zealand we are at 3 confirmed cases, one of which was an individual who recently return from Iran. There is a case that there are more yet undetected cases, but I suspect we are currently talking about dozens of people rather than "widespread" outbreak. Most of the initial testing was sent to Australia and I'm not sure what capability NZ medical labs have yet and I doubt they are sequencing much. There is a chance that the Iranian was sequenced, with media referred to a "deeper" testing protocol after initial tests where negative.
- DyslexicAtheist 7y agoseems the Iranian government can't be trusted with the information they put out there in any circumstance: https://twitter.com/aliostad/status/1233668766124400640 https://twitter.com/aliostad/status/1233668766124400640 https://news.ycombinator.com/item?id=22457840 https://news.ycombinator.com/item?id=22457840
- psim1 7y agoChina always wants to be seen as a global influencer. Now they certainly are, to their great shame.
- dang 7y agoPlease do not post nationalistic flamebait, or any flamebait, to Hacker News. https://news.ycombinator.com/newsguidelines.html https://news.ycombinator.com/newsguidelines.html
- julienchastang 7y agoHas anyone tried to play the "movie" in the map. As I understand it, there has already been back-propagation of the virus from North America back to Asia. Can anyone confirm this is actually the case? Edit: On a separate note, this tool is one of the best "web meets science" websites I've ever seen. Really nice work for those involved in its creation.
- mmmrtl 7y agoThat backprop is most likely an artifact of a sample on a lineage being sequenced in North America, but the lineage also continues in Asia, where that lineage had not been sequenced yet. As in, someone in Asia spread COVID-19 to two people, one of whom traveled to North American and had their virus sequenced. The lineage also continued in Asia and was eventually sequenced days/weeks later, so the inferred tree includes a North American sample seeming to give rise to more Asian infections.
- borkt 7y agoMy father in law is a pilot (currently not showing any systems virus) has been back and forth between China, Korea, HK (Guangzou most recently) at least 3 times since January. His last trip he was one of only 30 people in the hotel, being monitored every 30 minutes or so for fever while on a layover. His company is still scheduling him trips there, so I have no doubt with companies still requiring similar travel that it could have easily gotten back to China from someone who has been in the US over the past few months, now that we know it has been in the wild here for at LEAST 6 weeks.
- kire2345 7y agoSo am I reading this correctly that for example the strain in Lombardy Italy/CDG1/2020 (GISAID EPI ISL 412973) from 20.02.2020, has its origin in the strain collected in Bavaria Germany/BavPat1/2020 (GISAID EPI ISL 406862) from 28.01.2020. So the outbreak in Bavaria that was thought be contained somehow made its way to Italy? Or do these kind of interpretations not make any sense?
- cstejerean 7y agoThat seems like the right interpretation. The second case in Washington State appears to be a descendant of the first case on January 19, so we’ve also had about 6 weeks of undetected spread here from a case that was believed to be contained. It goes to show that “contained” doesn’t mean much when you only test people with symptoms but you know that the virus can spread asymptomatically. Early on the official guidance was that asymptomatic spread while possible is not playing a major role in global transmissions. Now we’re seeing the results of that position.
- consp 7y agoOr people having just a mild variant which does not cause fever and airway problems. If they think it is something else, or don't care, it is just as well a transfer risk as asymptomatic individuals and in my opinion at least as likely.
- all_blue_chucks 7y agoThis variant has killed two people already, so it's not too mild.
- borkt 7y agoIf it has only killed two (either one or both patients at a nursing home) after spreading undetected for 6 weeks, then there is reason to hope it is a less fatal variant.
- inciampati 7y agoAm I reading the documentation correctly, that you have to register at GISAID to download the sequences, and that you can't publish any part of the sequences in subsequent papers? Would it be problematic in any way to just release them under a CC-BY-SA-NC license?
- bionhoward 7y agoyou can download a bunch of sequences here (i filtered for lung and 'oronasopharynx'): https://www.ncbi.nlm.nih.gov/labs/virus/vssi/#/virus?SeqType_s=Nucleotide&VirusLineage_ss=Severe%20acute%20respiratory%20syndrome%20coronavirus%202,%20taxid:2697049&Isolation_csv=lung&Isolation_csv=oronasopharynx https://www.ncbi.nlm.nih.gov/labs/virus/vssi/#/virus?SeqType...
- jmartrican 7y agoCould it be the case that the least symptomatic variants of the virus will be the winning ones in the long run? I suspect that variants that do the most damage will be the ones quarantined and the ones that do least amount of harm (least symptoms) will spread faster. If so, will this virus just domesticate itself (or rather domesticated by humans filtering out those more deadly versions)? By domesticate I mean that it lives with us just like the common cold without causing a 10% dip in the stock market.
- emiliobumachar 7y agoThe common cold definitely destroys vast amounts of wealth day after day, we just don't see a dip because it's regular and predictable, that is, already priced in stock prices. As to the question, yes, that can well happen. Note that we could also argue that the common cold virus has domesticated us. We're its cattle.
- rolph 7y agoi wish i could say yes, but i cant. in the short term, if a virus can replicate itself and pass on to another host , that will in turn transmit the virus,,, that means a success. if debilitating symptoms occur after transmission occurs then those symptoms dont influence selection, this is what happens with asymptomatic incubation and infectious states. this type of virus uses RNA to carry its code, and a property of RNA is that it is much more error prone than DNA so accidental variations in the code occur quite frequently. A lot of these variants fizzle out but some of them will chance across an advantageous change in the code. so we see a virus that passes through populations and shifts its codeing by being error prone. this can result in dynamic swings between dominance of strains over time, and lead to innovations such as high communicability, or asymptomatic transmission.
- Symmetry 7y agoThat's normal. Diseases act as parasites on their hosts and its not in the best interest of a parasite to kill its host, or at least not quickly. That's what tends to make zoonotic diseases so dangerous, they've calibrated themselves to exist in an equilibrium with a host animal population but when they transfer to humans that balance is out of whack and they either are immediately wiped out by the immune system or they tend to overdo it and kill the host. COVID-19 seems to have "gotten lucky" in that it found itself with, from its perspective, a thankfully low lethality allowing it to spread widely. This can apply to bacteria even more than viruses, see https://aeon.co/essays/when-bacteria-kill-us-it-s-more-accident-than-assassination https://aeon.co/essays/when-bacteria-kill-us-it-s-more-accid... EDIT: I should say, though, that you shouldn't expect that to happen quickly. The selective pressure for COVID-19 to become less lethal is far less than it was for, say, MERS.
- united893 7y agoHad no idea that we could trace tiny differences so elaborately, this is amazing and gives me hope.