That Tencent transient webpage numbers looks about right when compared with the epidemiological model published by the guys an HK Univ Med School last week. We should be at around 1M infected but only 200K’ish with strong enough symptoms to be diagnosed. The 25K deaths for the whole county would be around 2.5% mortality rate. That HK model gives the Chinese epidemic reaching peak in late April. I have not seen any epidemiological models for the West yet but I would expect a two or three month lag with a lower peak.
For those wondering about treatment regime this is the clearest one published so far. The guy in Seattle who got infected…
Basically just the usual acetaminophen, ibuprofen and guaifenesin, with antibiotics so the patient did not pick up hospital inflections With some supplemental oxygen. And when it started turning into pneumonia they threw a shit load of antivirals at it for a day. Which seemed to work.
Long term future of 2019 nCoV will probably be something like this.
It seems the other corona viruses out there are the cause of a minority of common colds.
The reason why the Indian paper, now withdrawn, first got my attention was because of these two very odd paragraphs in one of the first characterizations of the virus.
" As a typical RNA virus, the average evolutionary rate for coronaviruses is roughly 10−4 nucleotide substitutions per site per year,
with mutations arising during every replication cycle. It is, therefore, striking that the sequences of 2019-nCoV from different patients described here were almost identical, with greater than 99·9% sequence identity. This finding suggests that 2019-nCoV originated from one source within a very short period and was detected relatively rapidly. However, as the virus transmits to more individuals, constant surveillance of mutations arising is needed"
" Phylogenetic analysis showed that bat-derived coronaviruses fell within all five subgenera of the genus Betacoronavirus. Moreover, bat-derived coronaviruses fell in basal positions in the subgenus Sarbecovirus, with 2019-nCoV most closely related to bat-SL-CoVZC45 and bat-SL-CoVZXC21, which were also sampled from bats.
These data are consistent with a bat reservoir for coronaviruses in general and for 2019-nCoV in particular. However, despite the importance of bats, several facts suggest that another animal is acting as an intermediate host between bats and humans. First, the outbreak was first reported in late December, 2019, when most bat species in Wuhan are hibernating. Second, no bats were sold or found at the Huanan seafood market, whereas various non-aquatic animals (including mammals) were available for purchase. Third, the sequence identity between 2019-nCoV and its close relatives bat-SL-CoVZC45 and bat-SL-CoVZXC21 was less than 90%, which is reflected in the relatively long branch between them. Hence, bat-SL-CoVZC45 and bat-SL-CoVZXC21 are not direct ancestors of 2019-nCoV."
Further studies should clear up these mysteries. Maybe. One odd feature about the Indian paper before it was shown to be wrong about the math was the pre-print site looked like it was being DDOS’ed. The Cloudflare behavior I saw was just like you see when a DDOS attack is detected. Maybe nothing to it, but still odd.
So I think we are looking at a rerun of the Asian flu epidemic of the late 1960’s. Those kind of numbers. With the most likely vector into the west being Australia. If the stories about the delay in shutting down flights for almost a week for the very lucrative foreign students from China to return are true. If looks like the hospital strike by medical staff in Hong Kong for the last two days to force the government to close the border has worked. So that should greatly reduce the spread into the west by that route.
The weibo page for CCTV will show you a lot of domestic TV news coverage. Using Chrome will give you a reasonable translation of the text. The expected positive story spin but not the old style heavy handed propaganda from the 1990’s and it will put very human faces on this huge humanitarian disaster.