This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Archive 1 |
An anonymous editor changed the "Avian" in the GISAID name to "All". While this makes sense, I can't find any evidence that the name was actually changed, so have reverted. Pol098 ( talk) 12:59, 19 June 2009 (UTC)
The GISAID logo seems to be updated with a gradient fill. I don't have experience loading new images, and I hope that's not a copyright issue, but I'll try to figure out how to do it on Commons. - AppleBsTime ( talk) 03:36, 5 June 2020 (UTC)
This topic is part of a series. For the series summary see [3] above.
Change the introduction from this:
Since its establishment as an alternative to sharing avian influenza data [1] via conventional public-domain archives, [2] [3] GISAID has been recognized for incentivizing rapid exchange of outbreak data [3] during the H1N1 pandemic [4] [5] in 2009, the H7N9 epidemic [6] [7] in 2013, and the COVID-19 pandemic [8] [9] in early 2020.
To this:
Since its establishment as an alternative to sharing avian influenza data [10] via conventional public-domain archives, [2] [3] GISAID has facilitated rapid exchange of outbreak data [3] during the H1N1 pandemic [4] [11] in 2009, the H7N9 epidemic [12] [13] in 2013, the COVID-19 pandemic [14] [15] and the 2022–2023 mpox outbreak [16].
most notably adding the mpox outbreak.
AncientWalrus ( talk) 12:11, 25 March 2023 (UTC)
@ AppleBsTime: Thanks for your contributions to this article. Since last December it has averaged over 200 views per day. That seems to me to be a handsome return for your work.
Is there a reasonably easy way for a mere mortal to obtain the number of unique sequences in the SARS-CoV-2 database?
Their website prominently reports the number of sequences submitted but not the number of unique sequences. This article currently (as of 2021-09-07) says, "by mid-April 2021, GISAID's SARS-CoV-2 database reached over 1,200,000 submissions". The Wikipedia article on Phylogenetic Assignment of Named Global Outbreak Lineages says, "The PANGOLIN web application has assigned more that 512,000 unique SARS-CoV-2 sequences as of January 2021."
Does that say that every other submission was unique?
Also, is it reasonable to say that the number of variants is roughly proportional to the number new cases? If yes, I estimate that the world has seen the probability of a new patient generating a new variant may be between 0.2 and 3 percent. I get the upper limit by considering that GISAID is not receiving submissions from a substantial portion of the world. Thanks, DavidMCEddy ( talk) 16:05, 7 September 2021 (UTC)
Harvey et al. (2021) claimed that between December 2019 and October 2020, the virus was "acquiring approximately two mutations per month in the global population." That's grossly misleading, because the mutation rate is not a function of time: It's proportional to the number of patients infected and spreading a disease. If the number of new cases per unit time is cut in half or by a factor of 10 or a million, the number of new variants per unit time will also be cut by approximately the same factor. We estimate that there is at least one new viable variant for each 600 cases, and it could be one for every 60 cases or less.[13]
The Wikipedia article on "Phylogenetic Assignment of Named Global Outbreak Lineages" claimed that more than 512,000 unique SARS-COV-2 sequences had been reported to open, international databases as of January 2021, all or nearly all of which could transmit the disease to another human. Unfortunately, no reference was given for that number. The actual number of unique sequences is almost certainly higher, because many parts of the world are not submitting sequences to the international database. The figure of almost 290 million cases by January 2021 came from the Wikipedia article on "COVID-19 pandemic cases", accessed 12 September 2021, which were extracted from the World Health Organization, "Coronavirus disease (COVID-19) Weekly Epidemiological Update and Weekly Operational Update" ( https://www.who.int/emergencies/diseases/novel-coronavirus-2019/situation-reports).
Hello, I have been taking some of my time on Wikipedia to improve this article, quite relevant during the COVID-19 pandemic, with newer sources and a more readable intro. I am noticing multiple edits made (and re-inserted) by IP addresses appearing to have the single purpose of editing Wikipedia exclusively about GISAID, yet no other subjects. The basis of these edits repeatedly seeks to convey the perception that GISAID's terms of access are "restrictive". Ironically, GISAID terms of use are not at all dissimilar to those of Wikipedia itself. Participating scientists are free to contribute or read from the database, just as long as they agree to appropriately acknowledge the contributors of the information they use. Contributors of data can freely choose whether they don’t care about any of their rights and deposit in public-domain archives, or whether they share in a transparent manner preserving some of their rights, and thus share with the public via GISAID. It's obviously a model that works -- Wikipedia has millions of articles under the Creative Commons Attribution license, and GISAID has over a million flu and about 50k genomic sequences of the virus causing COVID-19, contributed by thousands of laboratories under its usage license. Calling this a "restriction" is far less accurate than calling it "terms of use" or "regulating" how data are shared. I would like other contributors to consider this and respond, as I fear that this IP editor (or editors) is pushing an agenda and may be unlikely to form consensus. - AppleBsTime ( talk) 04:53, 20 June 2020 (UTC)
Much of this page appears (i) overly positive, to the extent that it serves as an advertisement for GISAID and (ii) uses sources that simply quote GISAID's positive description of itself. For examples, GISAID's web page says that it overcomes "disincentive hurdles and restrictions". The claim about disincentive hurdles is interesting, though vague. However, no example of "restrictions" is given. And yet, the current page repeats the claim that data sharing was "restricted". Additionally, the History section contains a list of "endorsements", which sounds like an advertisement, not a balanced description. Here are a list of other concerns:
Finally, user AppleBsTime has removed interesting facts, seemingly because they reflect negatively on GISAID. For example, the original signed letter calls for shared sequences to be deposited in public databases eventually, which would then allow scientists to share pre-publication data w/o being scooped while still not restriction post-publication data. But AppleBsTime removed this comment, even though it had a citation.
— Preceding unsigned comment added by 2620:0:691:4:0:0:0:1B ( talk) 23:28, 15 August 2020 (UTC)
There seems to be something of a disagreement in the 2 sections above. I just heard an excellent radio show/podcast on NPR, that gives a lot of info on this topic, but I wanted to check it out here. My general impression is that it considers the same disagreement as above with views from both sides.
Smallbones( smalltalk) 22:21, 30 May 2021 (UTC)
From the Duke researcher mentioned above: these articles from Nature and Science also talk about this disagreement:
I also note that http://www.nextstrain.org has begun offering two coronavirus tree reconstructions, one labeled "Latest Global Analysis - GISAID data", and one labeled "Latest Global Analysis - open data".
I am still confused about whether GISAID is trying to restrict PRE-publication data sharing or POST-publication data sharing. It sounds to me like it restricts both equally. 2603:6080:6502:E900:B7F:3D05:7E:3E7C ( talk) 04:46, 18 August 2021 (UTC)
Shu
was invoked but never defined (see the
help page).This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Archive 1 |
An anonymous editor changed the "Avian" in the GISAID name to "All". While this makes sense, I can't find any evidence that the name was actually changed, so have reverted. Pol098 ( talk) 12:59, 19 June 2009 (UTC)
The GISAID logo seems to be updated with a gradient fill. I don't have experience loading new images, and I hope that's not a copyright issue, but I'll try to figure out how to do it on Commons. - AppleBsTime ( talk) 03:36, 5 June 2020 (UTC)
This topic is part of a series. For the series summary see [3] above.
Change the introduction from this:
Since its establishment as an alternative to sharing avian influenza data [1] via conventional public-domain archives, [2] [3] GISAID has been recognized for incentivizing rapid exchange of outbreak data [3] during the H1N1 pandemic [4] [5] in 2009, the H7N9 epidemic [6] [7] in 2013, and the COVID-19 pandemic [8] [9] in early 2020.
To this:
Since its establishment as an alternative to sharing avian influenza data [10] via conventional public-domain archives, [2] [3] GISAID has facilitated rapid exchange of outbreak data [3] during the H1N1 pandemic [4] [11] in 2009, the H7N9 epidemic [12] [13] in 2013, the COVID-19 pandemic [14] [15] and the 2022–2023 mpox outbreak [16].
most notably adding the mpox outbreak.
AncientWalrus ( talk) 12:11, 25 March 2023 (UTC)
@ AppleBsTime: Thanks for your contributions to this article. Since last December it has averaged over 200 views per day. That seems to me to be a handsome return for your work.
Is there a reasonably easy way for a mere mortal to obtain the number of unique sequences in the SARS-CoV-2 database?
Their website prominently reports the number of sequences submitted but not the number of unique sequences. This article currently (as of 2021-09-07) says, "by mid-April 2021, GISAID's SARS-CoV-2 database reached over 1,200,000 submissions". The Wikipedia article on Phylogenetic Assignment of Named Global Outbreak Lineages says, "The PANGOLIN web application has assigned more that 512,000 unique SARS-CoV-2 sequences as of January 2021."
Does that say that every other submission was unique?
Also, is it reasonable to say that the number of variants is roughly proportional to the number new cases? If yes, I estimate that the world has seen the probability of a new patient generating a new variant may be between 0.2 and 3 percent. I get the upper limit by considering that GISAID is not receiving submissions from a substantial portion of the world. Thanks, DavidMCEddy ( talk) 16:05, 7 September 2021 (UTC)
Harvey et al. (2021) claimed that between December 2019 and October 2020, the virus was "acquiring approximately two mutations per month in the global population." That's grossly misleading, because the mutation rate is not a function of time: It's proportional to the number of patients infected and spreading a disease. If the number of new cases per unit time is cut in half or by a factor of 10 or a million, the number of new variants per unit time will also be cut by approximately the same factor. We estimate that there is at least one new viable variant for each 600 cases, and it could be one for every 60 cases or less.[13]
The Wikipedia article on "Phylogenetic Assignment of Named Global Outbreak Lineages" claimed that more than 512,000 unique SARS-COV-2 sequences had been reported to open, international databases as of January 2021, all or nearly all of which could transmit the disease to another human. Unfortunately, no reference was given for that number. The actual number of unique sequences is almost certainly higher, because many parts of the world are not submitting sequences to the international database. The figure of almost 290 million cases by January 2021 came from the Wikipedia article on "COVID-19 pandemic cases", accessed 12 September 2021, which were extracted from the World Health Organization, "Coronavirus disease (COVID-19) Weekly Epidemiological Update and Weekly Operational Update" ( https://www.who.int/emergencies/diseases/novel-coronavirus-2019/situation-reports).
Hello, I have been taking some of my time on Wikipedia to improve this article, quite relevant during the COVID-19 pandemic, with newer sources and a more readable intro. I am noticing multiple edits made (and re-inserted) by IP addresses appearing to have the single purpose of editing Wikipedia exclusively about GISAID, yet no other subjects. The basis of these edits repeatedly seeks to convey the perception that GISAID's terms of access are "restrictive". Ironically, GISAID terms of use are not at all dissimilar to those of Wikipedia itself. Participating scientists are free to contribute or read from the database, just as long as they agree to appropriately acknowledge the contributors of the information they use. Contributors of data can freely choose whether they don’t care about any of their rights and deposit in public-domain archives, or whether they share in a transparent manner preserving some of their rights, and thus share with the public via GISAID. It's obviously a model that works -- Wikipedia has millions of articles under the Creative Commons Attribution license, and GISAID has over a million flu and about 50k genomic sequences of the virus causing COVID-19, contributed by thousands of laboratories under its usage license. Calling this a "restriction" is far less accurate than calling it "terms of use" or "regulating" how data are shared. I would like other contributors to consider this and respond, as I fear that this IP editor (or editors) is pushing an agenda and may be unlikely to form consensus. - AppleBsTime ( talk) 04:53, 20 June 2020 (UTC)
Much of this page appears (i) overly positive, to the extent that it serves as an advertisement for GISAID and (ii) uses sources that simply quote GISAID's positive description of itself. For examples, GISAID's web page says that it overcomes "disincentive hurdles and restrictions". The claim about disincentive hurdles is interesting, though vague. However, no example of "restrictions" is given. And yet, the current page repeats the claim that data sharing was "restricted". Additionally, the History section contains a list of "endorsements", which sounds like an advertisement, not a balanced description. Here are a list of other concerns:
Finally, user AppleBsTime has removed interesting facts, seemingly because they reflect negatively on GISAID. For example, the original signed letter calls for shared sequences to be deposited in public databases eventually, which would then allow scientists to share pre-publication data w/o being scooped while still not restriction post-publication data. But AppleBsTime removed this comment, even though it had a citation.
— Preceding unsigned comment added by 2620:0:691:4:0:0:0:1B ( talk) 23:28, 15 August 2020 (UTC)
There seems to be something of a disagreement in the 2 sections above. I just heard an excellent radio show/podcast on NPR, that gives a lot of info on this topic, but I wanted to check it out here. My general impression is that it considers the same disagreement as above with views from both sides.
Smallbones( smalltalk) 22:21, 30 May 2021 (UTC)
From the Duke researcher mentioned above: these articles from Nature and Science also talk about this disagreement:
I also note that http://www.nextstrain.org has begun offering two coronavirus tree reconstructions, one labeled "Latest Global Analysis - GISAID data", and one labeled "Latest Global Analysis - open data".
I am still confused about whether GISAID is trying to restrict PRE-publication data sharing or POST-publication data sharing. It sounds to me like it restricts both equally. 2603:6080:6502:E900:B7F:3D05:7E:3E7C ( talk) 04:46, 18 August 2021 (UTC)
Shu
was invoked but never defined (see the
help page).