Talk:Wuhan coronavirus outbreak

Origin of 2019-nCoV?
The genome of 2019-nCoV has been claimed to come from four different sources:
 * 1) Bat coronavirus similar to Bat-SL-CoVZXC21 or Bat_SL-CoVZC45
 * 2) Spike glycoprotein gene from human SARS
 * 3) pShuttle-SN vector used in labs for splicing the genome
 * 4) HIV inserts at the tips of the spikes in the spike glycoprotein

Points 2 and 3 are wrong!

I analyzed the two claims made about inserts. James Lyons-Weiler claims that the 2019-nCoV virus has a unique sequence about 1,378 bp (nucleotide base pairs) long that is not found in related coronaviruses. He published the sequence online. He also claims that the sequence also contains the pShuttle-SN expression vector.

I ran the sequence through an online DNA to protein translator. Reading it in reading frame 2 (i.e leaving out the first "c") gives this amino acid sequence.

SVLHSTQDLFLPFFSN VTWFHAIHVSGTNGTKRFDNPVLPFNDGVYFASTEKSNIIRGWIFGTTLDSKTQSLLIVNNATNV VIKVCEFQFCNDPFLGVYYHKNNKSWMESEFRVYSSANNCTFEYVSQPFLMDLEGKQGNFKNLRE FVFKNIDGYFKIYSKHTPINLVRDLPQGFSALEPLVDLPIGINITRFQTLLALHRSYLTPGDSSS GWTAGAAAYYVGYLQPRTFLLKYNENGTITDAVDCALDPLSETKCTLKSFTVEKGIYQTSNFRVQ PTESIVRFPNITNLCPFGEVFNATRFASVYAWNRKRISNCVADYSVLYNSASFSTFKCYGVSPTK LNDLCFTNVYADSFVIRGDEVRQIAPGQTGKIADYNYKLPDDFTGCVIAWNSNNLDSKVGGNYNY LYRLFRKSNLKPFERDISTEIYQAGSTPCNGVEGFNCYFPLQSYGFQPTNGVG

The sequence is identical to the sequence published in the now withdrawn Indian paper. It corresponds to positions 46 to 504 in the full protein (positions 50 to 508 in the alignment table). The sequence is a close match to the SARS spike protein. There is no place to fit in the pShuttle-SN sequence, unless the pShuttle-SN sequence itself mimics or contains the SARS spike protein. James Lyons-Weiler now admits that this is in fact the case.

James Lyons-Weiler also claimed that the spike glycoprotein of 2019-nCoV is most similar to the SARS spike protein and not to the SARS-like coronavirus in bats. This is not true. The article in The Lancet records a 80.2% match between the spike protein of 2019-nCoV and Bat_SL-CoVZC45 but only 76.2% between 2019-nCoV and SARS. This level of similarity is also shown in the sequence alignment presented in the Indian paper.

The Indian paper makes the mistake of comparing 2019-nCoV to SARS, when in fact it is most related to SARS-like bat virus Bat_SL-CoVZC45 or Bat_SL-CoVZXC21 or some common ancestor. This however does not invalidate their results. There are a near infinite number of current and past viruses in the wild that are ever closer ancestors to 2019-nCoV. Comparing 2019-nCoV to SARS is not fundamentally different to comparing to a bat coronavirus or to some yet to be discovered closer relative.

The paper should be rewritten by comparing 2019-nCoV to Bat_SL-CoVZC45. It would also be interesting to know if the insets match the DNA sequence of the database of known HIV genomes and not only the amino acid sequence. I would do the comparison myself, but I do not yet have access to the 2019-nCoV or Bat_SL-CoVZC45 full genomes.

The spike protein of SARS is 1255 amino acids long, nine more than in Bat_SL-CoVZC45. A transformation from SARS to 2019-nCoV removes four amino acids and adds 22 for a total of 1273. A comparison of Bat_SL-CoVZC45 to 2019-nCoV should see inserts of 31 to 35 amino acids or 93 to 105 nucleotides.

I still think the link to HIV sequences is statistically meaningful. (continued.)


 * http://www.tiem.utk.edu/~gross/bioed/webmodules/aminoacid.htm
 * https://en.wikipedia.org/wiki/HIV
 * https://en.wikipedia.org/wiki/Structure_and_genome_of_HIV
 * https://en.wikipedia.org/wiki/Envelope_glycoprotein_GP120

-- Petri Krohn (talk) 03:57, 5 February 2020 (UTC)

Synthia?
This is probably a lunacy, but I'd like to look into and have a record on it anyway.
 * New version of the appearance of coronavirus COVID19, February 20, 2020, via-midgard.com (randomly walked into from some other page), robo-translation
 * Synthia (Mycoplasma laboratorium) is an artifical modification of mycoplasma with a synthetic genome which can reproduce itself, created by Craig Venter and Venter Institute in 2010 (true)
 * Mycoplasma pneumonia may cause atypical pneumonia in humans and post-pneumonia complications, eg autoimmune disease (true, but Synthia was based on Mycoplasma genitalium; how much they differ, I cannot say)
 * There was a Deepwater Horizon accident on an oil rig in the Gulf of Mexico in 2010, resulting in a massive oil spill (true)
 * Artificial oil-eating bacteria was used to combat the oil spill -some sort of biodegradation was used, details, and how artificial were they, are sketchy (eg here); also some indigenous microorganisms contributed to oil decomposition.
 * Synthia was that artificial oil -eating bacteria. - I not aware of any evidence that Synthia is oil eating. That part seems to be coming from conspirology websites making similar claims. This does not exclude, per se, that some artificial bacteria was created or tried to reduce oil spill, however I could not find any reliable information that this was the case
 * That bacteria attacked humans as well. In view of the above, it is not supported, however there were reports of various health and breathing problems in human population in the area, in particular due to algae (red tide), and other reasons. Official Texas respirotary data for 2010 and 2011 are pretty confusing. 9115 deaths from respiratory problems of all sorts in 2011, 35.5 per 100,000 people. In 2010, 3,013 deaths, 12.0 per 100,000, from pneumonia and influenza. There may be increased rate in 2011, but data are not in the same format to be certain.
 * As the story goes, fish ate that bacteria and eventually surfaced in Wuhan in 2019, to be a source of a chimera with coronavirus. That is not evidenced and probably is a long shot, but I cannot quickly refute it with certainty; and I have not done sequencing type stuff in practice before. --Resup (talk) 12:10, 22 March 2020 (UTC)

Coronavirus with time machine?
I have reviewed the article preprint quoted in this article by China Global Television Network. It is bogus!


 * New study further proves Wuhan seafood market not the source of COVID-19
 * ''A recent study conducted by a group of Chinese scientists have found further genome evidence to prove that the seafood market in Wuhan is not the source of the novel coronavirus – a claim first made in a paper published on The Lancet.
 * ''The study, led by researchers from Xishuangbanna Tropical Botanical Garden of Chinese Academy of Sciences, South China Agricultural University and Chinese Institute for Brain Research, was published on ChinaXiv on Saturday in a pre-print version without peer review.
 * ''Per the study, genetic data suggests the virus was introduced from elsewhere and had already circulated widely among humans in Wuhan before December 2019, probably beginning in mid to late November.
 * ''The crowded seafood market facilitated the virus transmission to buyers and spread to the whole city on a large scale in early December 2019, corresponding to the estimated population expansion time, the study shows.
 * ''Researchers collected the genome-wide data from 93 new coronavirus samples shared on the GISAID EpiFlu, an international database that stores information about influenza virus, to study the evolution and human-to-human transmission of the virus over the past two months.

Did haplotype H3 come before H1? The article preprint claims so:


 * Decoding evolution and transmissions of novel pneumonia coronavirus using the whole genomic data
 * ''H1 and its descendant haplotypes from the Hua Nan market should be derived from the H3 haplotype, which was not linked to the market.

The phylogenetic network alone cannot determine what was the ancestral haplotype. Mutations can have happened both ways. There are three ways of deriving the ancestral node from a phylogenetic network: The ChinaXiV paper is doing this wrong. The known cases of H3 appeared one month later than H1. This is evident from the haplotype timeline included in the article.
 * 1) Date the appearance of each haplotype.
 * 2) Follow known infection patterns.
 * 3) Compare genomes to known distant ancestor (bat-RaTG13-CoV).

Neither have they demonstrated the link to link to bat-RaTG13-CoV through their hypothesized haplotype mv1. Choosing another hypothesized haplotype could as well have produced a link to some other SARS-CoV-2 haplotype. -- Petri Krohn (talk) 14:05, 24 February 2020 (UTC)

Italy likely to have 60,000 COVID-19 infections already
When Wuhan and Hubei were locked down on January 23, 2020 there were only 830 coronavirus infections in all of China. As of today there are 67760 cases in Hubei, almost a 100-fold increase! Yet the lockdown seems to have been effective as the epidemic is now over.

When northern Italy was put under lockdown on March 8, 2020 there were 7,375 cases in Italy, with a daily growth rate of 25%. From these numbers it is possible to estimate the total number of people in Italy already infected with the COVID-19 coronavirus.

The growth rate in China on and around January 23rd was about 50% per day. Going from 830 to 67760 would take about 10.86 days. This number is related to the average incubation period.

Staring from 7,375 with a daily growth rate of 25% for 10.86 days results in a 11.28-fold increase or a total of 60,613 cases.

This estimate is based on the assumption that the daily growth rate at the time of the lockdown was twice as high in China as in Italy. Italians may have already prepared for the coronavirus and practiced better hygiene. The figure 50% for the Chinese growth rate is a rough estimate. Averaging over 10 days before and after January 23rd gives a daily growth rate of 47.6%.

It may also be that the growth rates are similar. Comparing the Italian and Chines numbers side-by-side gives similar growth rates for the same number of patients, the main difference being the Italy started the quarantine and lockdown 6 days later and with nine times the number of cases. If so, the total number of COVID-19 cases in Italy may reach half a million.

-- Petri Krohn (talk) 22:22, 10 March 2020 (UTC)

Carlo Urbani
Carlo Urbani is credited with ending SARS epidemic in Vietnam by introducing strict quarantine measures. He got ill himself, documented his illness, and died on arrival to a conference in Bangkok in 2003. It is said that acute complications from the pneumonia had quick onset of about 3 hours --Resup (talk) 05:03, 15 March 2020 (UTC)

Russian chief pulmanologist interview

 * Interview with Alexand Chucalin, RT, titled "how to treat COVID-19", 25 min.

This is a long interview, with the essence appear to be that this a rather meek virus by itself and is not the killer itself; the real killer is bacterio-viral complications which may develop in some patients. The end stages in that case is Non-cardiogenic pulmonary edema, which require being placed on ventilator (if enough are available). While on ventilator, suppressed immunity may develop and rather mundane microorganisms are the eventual killers in that case.

Non-cardiogenic pulmonary edema ''develops both with direct lung damage (inhalation of toxins / toxic gases, aspiration of the contents of the gastrointestinal tract, water, blood, burns, pneumonia), and with extrapulmonary diseases (shock, severe polytrauma, sepsis, massive blood loss, pancreatitis / pancreatic necrosis, blood transfusion, uremia) is much less common. Another syndrome manifested by non-cardiogenic pulmonary edema and acute respiratory failure is ARDS - acute primary / secondary respiratory distress syndrome (synonymous with wet lung, shock lung). The development of this type of edema is mainly due to a violation of the function / structure of the alveolocapillary membrane.

''According to modern concepts of primary ARDS, the damaging factor directly affects alveolocytes, a surfactant that leads to damage to the alveolar epithelium and impaired integrity of the alveolocapillary membrane, resulting in transfusion of the liquid component of blood into the alveolar spaces. With secondary ARDS, the extrapulmonary factor is the basis for it, which forms against the background of the inflammatory response syndrome in the body (sepsis, bacteremia). Edema of this type (shock lung) does not obey the laws of fluid transport and does not depend on the level of hydrostatic pressure.

''Its peculiarity is a sharp decrease in ventilation / oxygenation of the body, which causes congestion in the lungs, oxygen deficiency of brain and heart tissues with the rapid development of life-threatening conditions. With inadequate / untimely assistance, mortality reaches 65-80%.

--Resup (talk) 05:22, 15 March 2020 (UTC)


 * "How to treat with coronavirus infection" - Cassad coverage of the interview, March 14, 2020

“The disease itself has at least four such outlined stages: the first stage is viremia. A harmless cold, nothing special, seven to nine days approximately in this interval, ”said Chuchalin. According to him, from the ninth to the 14th day of the disease, the situation "changes qualitatively," because it is during this period that viral-bacterial pneumonia forms. “After the epithelial cells of the body are damaged, microorganisms and bacteria are colonized in this anatomical space of the airways,” he explained. According to him, the doctor should "show his skill" to identify the disease in the initial stages. “If this situation is not controlled and the disease progresses, then more serious complications come, we call it acute respiratory distress syndrome, shock, a person cannot breathe on his own. To be precise, we call this non-cardiogenic pulmonary edema, ”said the specialist. According to him, pulmonary edema can be treated with a mechanical ventilation machine (mechanical ventilation). “If a person suffers this phase, then the immunosuppression caused by the defeat of acquired and innate immunity becomes fatal and the patient joins such aggressive pathogens as Pseudomonas aeruginosa, fungi”

Bill Ackman interview

 * Bill Ackman interview, YouTube, March 18, 2020
 * '' "Pershing Square Capital CEO Bill Ackman makes a plea to President Trump to shut the U.S. economy down." --Resup (talk) 20:15, 18 March 2020 (UTC)

Treat COVID-19 with immunosuppressors?
I suggest that immunosuppressants are tried as a treatment for the pneumonia associated with the new coronavirus disease. This idea comes from combining five pieces of information and speculation.


 * 1) Despite over a decade of efforts no one has been able to develop a vaccine for SARS or any other coronavirus. The problem is that the test subjects vaccinated (usually laboratory mice) exhibit a too strong immunoreaction to the virus, sometimes leading to a lethal cytokine storm.
 * 2) COVID-19 patients become highly infectious before symptoms appear, but are no longer infectious as the disease progresses. This shows that the immune system is working and able to kill all viruses outside cells but viruses may still survive inside the lung cells.
 * 3) All symptoms of COVID-19 are identical to the mysterious "vaping illness" that caused and epidemic in the United States in the fall of 2019. In both diseases CT scans show "ground-glass opacities" in the lungs. (The cause of "vaping illness" has never been properly established. It is my suspicion that it is actually caused by a strain of the SARS-CoV-2 virus.)
 * 4) Patients with hypersensitivity pneumonitis, an immune system disorder that affects the lungs show "ground-glass opacities" in CT scans.
 * 5) Immunosuppressors have been used successfully in treating "vaping illness".
 * 6) * Outbreak of Electronic-Cigarette–Associated Acute Lipoid Pneumonia — North Carolina, July–August 2019
 * ''All five patients improved clinically within 24–72 hours after initiation of intravenous methylprednisone (120 mg–500 mg daily). All five patients survived and were discharged home on a taper of oral prednisone.

-- Petri Krohn (talk) 21:17, 20 March 2020 (UTC)

After I wrote the above I noticed that a letter to The Lancet says essentially the same thing. -- Petri Krohn (talk) 23:12, 20 March 2020 (UTC)
 * COVID-19: consider cytokine storm syndromes and immunosuppression - Puja Mehta & al., The Lancet, March 16, 2020
 * ''Accumulating evidence suggests that a subgroup of patients with severe COVID-19 might have a cytokine storm syndrome. We recommend identification and treatment of hyperinflammation using existing, approved therapies with proven safety profiles to address the immediate need to reduce the rising mortality.
 * ''Predictors of fatality from a recent retrospective, multicentre study of 150 confirmed COVID-19 cases in Wuhan, China, included elevated ferritin and IL-6 suggesting that mortality might be due to virally driven hyperinflammation.

My take: there are indeed autoimmune-type forms of pneumonia (or complications after it); whether or not this is the case here is one of the questions to be answered, but apparently present view that it is not, Supposedly autoimmune response will create markers picked by tests, like blood tests, and we would know about it fairly quickly. In presentation of Russian chief pulmanologist, the impression is that ARDS which develops is caused by weaken immune response, not by over-response (but this may need more clarity). He does mention vaping matter, giving an impression that it is not understood, and commenting that 'they die once they are placed on ventilator'; he does not claim that vaping illness and COVID-19 have more in common. -- Resup (talk) 07:56, 21 March 2020 (UTC)

Herd immunity

 * The U.K. backed off on herd immunity. To beat COVID-19, we’ll ultimately need it. - National Geographic, March 20, 2020 --Resup (talk) 10:23, 21 March 2020 (UTC)

Basic reproduction number R0 for COVID-19 is infinite!
I have been trying to find a paper that would give estimates for the all three parameters of the epidemic, the basic reproduction number R0, the daily growth rate and the serial interval. So far I have not found any.

The growth rate in most Western countries is ten-fold every week or one thousand fold every month. The first cases of the Spanish epidemic were discovered on February 25. Now, 31 days later Spain has 57,786 laboratory confirmed cases.

The best estimates for the values are in the latest study by the London‌ Imperial College. Instead of estimating a R0 they give graphs for R(t) and its variation over time. The initial values are somewhere between 7 and 10. The paper also states a mean serial interval of 6,48 days.

An estimate for the future development of the pandemic in Finland by Finnish health officials is not sure what the R0 value for COVID-19 is. They claim that R0 in Lombardy was initially "little over 3" and give two prognosis for Finland based on R0 values of 1.6 and 1.8. The whole estimate is worthless crap!

There is no R0 value for COVID-19. Any real time value of R(t) is a function of social distancing measures. The best surrogate for a R0 value is infinity (∞). Everyone you meet will be infected! Everyone in the same room will be infected. If someone is infected in a concert hall with 1000 people, all one thousand are likely to be infected. The only way to decrease R(t) is by limiting the number of people you meet while infectious.

But even the stringiest social distancing measures cannot avoid a pandemic. As long as people meet and interact, everyone will eventually be infected. The only way of avoiding a pandemic and pushing R(t) under 1 is to isolate infected people before they become infectious. This requires tracing contacts and placing them all under quarantine.

-- Petri Krohn (talk) 23:57, 26 March 2020 (UTC)

Petri, there is another Wuhan-based paper which appears to me (on a very brief glance) more informed and more informative. It has supplements discussing parameters as well as SEIR -type model they use. They say that R(t) you worry about dropped to below 1 after most stringent of measures taken, centralized quarantine. I suppose that means quarantine for family members and all established contacts, totalitarian sort of way (And presumably not something likely to happen in Western pseudo-democracies, a China-Russia advocate would say).

And it was above 1 with only social distancing/isolation, in an earlier period (But noted, my understanding of this being claimed is mostly based on a Russian summary, not a particular place in the paper, which I looked into but in 'turning pages' mode as I could not (yet?) find time to do better).

My understanding is that R(t) (total number of directly ever infected by 1 person, with t dependence coming from changing preventive measures) is finite because a person is infectious for a limited time (recovering, getting isolated, or dying after a while); only direct infections (presumably) are counted, not via another party; and this is indeed affected by social interactions mode but an average (in some sense) is taken.

I also note that in SEIR model important parameter is not R(t) but infection rate parameter (I gather called b in paper I quote) in the dynamical system saying essentially that d I/dt = b/N S* I + other terms, d S/dt = -b/N S* I + other terms, S susceptible, I infected, N population size (I am yet to ascertain fine details here). Recovery of I is modeled by terms including dI/dt = - gamma I (+other terms). R(t) is then computed, taken into account this dynamics, and there is a formula somewhere in supplementary part of the paper I link. (Noted, comments written in a hurry as I do not really have much time; so details could be inaccurate; paper(s) need to be consulted for the accurate account).

I also note that available data is pretty 'dirty', for example in Italy number of deaths of people who had COVID is announced, but that does not mean COVID was the main cause of the death (if they would do this during a normal flu epidemic, results perhaps would appear quite dire too) --Resup (talk) 23:59, 27 March 2020 (UTC)


 * I have read that paper. They state an R0 of 3.88 but it is impossible to confirm, as they do not state a value for the serial interval or the growth rate they use. (The growth rate in the US is something like ten-fold in 8 days.) -- Petri Krohn (talk) 09:32, 28 March 2020 (UTC)