2021-01-19 00:25
Edward C. Holmes
"1. Yes, deletion at 69,70. 2. Yes. no deletion but HL at 69,70. Note human WH1 is HV at 69,70. 3. GD pangolin-CoV one is actually more genetic distant to human WH1 than than GX pangolin-CoV does. It is not deletion but with TK at 69,70. 4.For the existence of two copies in either GX-P2V (isolate) or GX-P4L (lung original sample), I have checked the read assembly, none of them contain both copies."
```GISAID did indeed acknowledge a) the anniversary of learning that the respiratory illnesses reported in Wuhan were not caused by an influenza virus, but by a previously unknown human coronavirus, and b) the anniversary ofGISAID releasing the first whole-genome sequences and metadata. The January 10th date is not a "mistake". Rather, reports that those genomes were first made available on GISAID on a later date are wrong. Ever since GISAID went live in May 2008, whenever there is word of a noteworthy event in which influenza-like symptoms are noted, GISAID goes into "first-responder-mode" and awaits feedback from its colleagues involved inthe investigations. Last January was no exception. On January 8th 2020 at 3:38am PST, we received a message from George Gao, who I copy here, saying: "It's a novel coronavirus". As you know, it is typical that GISAID receivessuch an alert when a novel strain appears, as in 2009, when GISAID was advised by the US CDC of H1N1p, and in2013, when the China CDC gave GISAID a heads up on the HPAI H7N9, and on numerous other occasions when itseemed an outbreak could occur. In almost all cases when we are alerted, GISAID receives the first data generated between 36-48 hours later, well prepared to process these data.George's alert gave us a chance to prepare, and gave our technical team time to set up an HTML area. We received the first 3x genomes and metadata in the afternoon of January 9th at 3:29 pm PST (2020-01-09 23:29 UTC). Two ofthese genomes, HB-01 and HB-05, were released on GISAID after processing, a little more than an hour later, at4:41 pm PST and 4:44 pm PST (2020-01-10 0:41 UTC and 2020-01-10 0:44 UTC), respectively.GISAID understood at that time that it would be dealing with only a few hundred cases, yet by the time you and Ispoke on January 17th, it was apparent to us both that this outbreak would likely see significantly more cases. Our GISAID technical team was immensely grateful for the technical advice you provided over the next several days, which helped them scale up and steadily improve the platform. You and I had subsequent interactions over the next few months. For example: on March 10th, when we discussed the inclusion of genomes with low coverage; on March 13th, when we debated which reference sequence from the early Wuhan cluster would be most suitable for use by GISAID, and on April 23rd, when you introduced your lineage naming system (which GISAID subsequently integrated).During all of our interactions, you never questioned that the first genomes were made available on GISAID onJanuary 10th, even though that date was prominently displayed on the website and in the database, the entire time that our discussions were taking place. Wouldn't you have said something if our date was wrong? Not even Prof. Holmes, who wrote to my colleague on January 20th, passing on an update request of metadata forBetaCoV/Wuhan-Hu-1/2019, took issue with the dates of the first genomes, clearly visible in the database.To be clear Andrew; it is certainly not my intention to challenge your statement about the date and time, that Prof. Holmes posted to your discussion forum on January 11th. I have the utmost respect for your scientific knowledge. The accuracy of reporting on GISAID during 2020 left much to be desired, to put it mildly. Some of what was published, even by reputable outlets, was a disservice to science. Just look at what the South China Morning Post(SCMP) wrote on January 11th, 2020: "The relevant institutions have also uploaded the genetic sequence onto one online GenBank (GISAID). GISAID is cross-checking the information and will publish it upon completion." When their article was published, the first genomes were already available on GISAID - which the SCMP should have known, because the article used the first image of the coronavirus from the GISAID homepage, the caption of which stated "China releases genetic sequence of the newly discovered coronavirus from Wuhan."It is disheartening that, nearly a year later, inaccurate -- and conflicting -- statements regarding when and where the first genomes were made available are still being disseminated. Two well-known outfits recently publishedconflicting accounts. Science News stated: "10 January Genetic sequence posted to virological.org", while Nature News states that the first sequence was posted on the website virological.org on 11 January 2020. Neither news organization had done even the most basic research, and did not realize that there were discrepancies in reports ofwhen and where the first coronavirus sequences were released to the world.I am sure we both agree this is not a competition. GISAID was not established for fame or compliments. What matters, is that we aid public health by incentivizing and encouraging data sharing in crises like the one we are currently experiencing, so that those developing medical interventions -- whether vaccine or diagnostics to mentiona few -- can proceed with confidence in the quality of the data they rely on.```