Talk:Variants of SARS-CoV-2

P1 > Assessment source

• Freitas, André Ricardo Ribas; Lemos, Daniele Rocha Queiróz; Beckedorff, Otto Albuquerque; Cavalcanti, Luciano Pamplona de Góes; Siqueira, Andre M.; Mello, Regiane Cristina Santos de; Barros, Eliana N. C. (2021-04-19). "The increase in the risk of severity and fatality rate of covid-19 in southern Brazil after the emergence of the Variant of Concern (VOC) SARS-CoV-2 P.1 was greater among young adults without pre-existing risk conditions". medRxiv: 2021.04.13.21255281. doi:10.1101/2021.04.13.21255281.

Delta mortality: CFR in England not applicable

See also: Talk:SARS-CoV-2 Delta variant § Mention of CFR in second paragraph of article

Overview table

It would be helpful if a row for the original virus could be added to the top of the table. I don't have the medical qualifications to do that. Lay people may assume the Alpha variant is the original virus unless they read the article carefully. Ghostofnemo (talk) 02:25, 22 July 2021 (UTC)

The table describes changes from the original variant and those similar to it. Adding a line for the original variant only makes sense if we describe absolutes rather than relatives, such as reproduction number (R₀ or R_t) rather than changes in transmissibility. So far, most of the data are being reported relative to the original variant because it is the one most governments have prepared for. Of course that will likely change. In the UK, for example, most reports now relate to Alpha because it has surpassed all other strains that behaved like the original. Soon, unfortunately, the reporting will likely establish Delta as the main reference. --Fernando Trebien (talk) 20:42, 22 July 2021 (UTC)

@Ftrebien: Would you consider a supplementary table include the absolute values to be a violation of the policy against original research? I'm not sure about that, which is why I'm hesitant to add a supplementary table. CessnaMan1989 (talk) 01:43, 5 November 2021 (UTC)

@CessnaMan1989: After this question was raised, I found and added a better reference with values relative to the original variant, so I don't think this is a problem anymore. But suppose we were still back on 22 July. The problem is that we would be composing two somewhat uncertain measurements to arrive at a new and less certain measurement. We should ask ourselves why researchers are publishing values relative to variants currently in circulation, rather than absolute values that would be comparable to the original variant. The researchers from this new source I found had direct access to the data to recalculate the confidence intervals, we in Wikipedia do not have that access and if we tried to do the same, it would be the same as taking the role of those researchers. So, yes, I think this would qualify as original research. The best we could do is multiply the two estimates, but then we would be unable to provide a confidence interval, so the reader wouldn't know how accurate the composite estimate is. That might be acceptable, but it would be better to add a note explaining where the composite value came from. --Fernando Trebien (talk) 12:26, 5 November 2021 (UTC)

Medrxiv sourcing and technical DOI issues

@Jaredroach: I fully understand medrxiv is not a proper Wikipedia reference. I'm glad you've decided to remove medrxiv references which are not based on any peer review. However, one of the articles you removed is actually also published in a reputable journal (Cell (journal)), and should have been updated instead of removed.

The point of my edit was to add in a DOI pointing to the peer reviewed paper. Before my edit, the article had a DOI pointing to the medrxiv preprint, but not the actual paper. Unfortunately, the Template:cite journal does not have a parameter for medrxiv preprints within the same citation, so I moved the preprint link to a separate template within the same footnote to keep the source accessible, since the publisher article is paywalled. I'm not adding any references here, just clarifying the DOI of the authoritative source.

I think your revert was based on a misunderstanding. Finding an article on medRxiv doesn't mean it's *not* peer reviewed, it's just not a sufficient evidence of peer review. 〈 Forbes₇₂ | Talk 〉 22:10, 2 October 2021 (UTC)

That's fine. But in no case should there be any need to reference medrxiv. If an article in medrxiv is published later in a reputable source, one should reference the reputable source. One should not reference medrxiv "pending" the availability of that reputable source. I think it is correct for one editor to remove the medrxiv source, and for another editor to add the new source. The first editor might not easily have the info for next source, nor should they feel obligated to find it before deleting the medrxiv source. After removing the medrxiv source, the material may be unsourced. A "citation needed" tag can be added, and eventually the material can be deleted or someone can add the new reference. Jaredroach (talk) 22:16, 2 October 2021 (UTC)

I agree with you that using preprints as "pending" sources should never be done, and any non-peer reviewed sources used as a reference should be deleted. I'd just point out that arxiv/biorxiv/medrxiv will link from the preprint directly to the published version if it exists, so it's a one-click check to see if the reference is salvageable. I assume you're just objecting to standalone medrxiv references as bad, and not any and all preprint links? Adding a link to the preprint is accepted practice on Wikipedia as long as the underlying peer-reviewed paper is also linked in the same citation. I three options to add the missing peer-reviewed link to the citation:

1. replace the DOI link to the preprint with a DOI link to the reviewed paper, but keep the preprint accessible in a nearby template (what I did before revert)

2. replace the DOI link to the preprint with a DOI link to the reviewed paper, remove the link to the preprint entirely (I would object if medrxiv was the only free source, but since there's a PMC link this doesn't matter much)

3. update the cite journal template so we can include both the DOI link to the paper and DOI link to the preprint in the same template.

I'd say option 3 is ideal, but it would take some time since cite journal is a highly visible, fully protected template. What do you think? 〈 Forbes₇₂ | Talk 〉 23:26, 2 October 2021 (UTC)

Absent explicit assurance that the content of the preprint is identical with the content of the peer-reviewed (published) version, they should not be considered equivalent. Peer review often results in revisions (it's not simply an approval process). — soupvector (talk) 23:42, 2 October 2021 (UTC)

That's a very technical distinction to make. In my experience comparing peer-reviewed papers to preprints, preprints verus published paper end up pretty similar, so I usually prefer to link to both in the case of a paywall. I agree the published one is the authoritative source. I think its a mistake to remove preprint links for peer-reviewed papers, but I've gone ahead and removed the preprint doi altogether in this one case so at least the citation template is correct. 〈 Forbes₇₂ | Talk 〉 01:59, 31 October 2021 (UTC)

No idea how easy it is to update MedrXiv preprints, but updating ArXiv preprints to match the final accepted version (in content, not format, font, etc) is standard practice. This guarantees green open access. Any non-trivial changes in the review process are normally put in the final versions of the preprints, because it's a lot easier to check ArXiv preprints than waste time trying to guess which journals are non-paywalled. Moreover, some papers have mathematical errors introduced by proofreaders that the authors fail to pick up when double-checking the proofs, in which case the version-of-record is wrong and the final preprint is right. The peer review process statistically improves the quality of research papers, but hiding or removing preprint information generally functions as a block against access to verifiable information, especially with journals that are still paywalled. Caveat: the actual practice may vary across disciplines - here I'm talking of astronomy/cosmology - I can't judge MedrXiv. Boud (talk) 06:51, 26 November 2021 (UTC)

"Positive" vs. "Negative" vs. "Neutral" mutations

Could somebody further explain to me what the video in the article means by "positive mutations" and "negative mutations"? What makes a mutation positive or negative? CessnaMan1989 (talk) 01:48, 5 November 2021 (UTC)

B.1.1.529

Just saw news of this breaking over the last couple of hours. It's currently listed on the WHO tracking page as a variant under monitoring. I'm not sure what counts as a RS in this topic area yet, I note however that Axios have stated that it's ben found so far in Botswana, South Africa, and Hong Kong. Reuters state 100 specimens in South Africa, and confirm what Axios state for Botswana and Hong Kong but state the Hong Kong case is a traveller from South Africa. Sideswipe9th (talk) 18:19, 25 November 2021 (UTC)

I've added the basic info from the WHO tracking page. Not sure what to cite for more detailed information in the notes section however. Sideswipe9th (talk) 18:25, 25 November 2021 (UTC)

Thanks User:Sideswipe9th for making the addition. I think the sheer number of mutations on the spike protein alone is the most concerning part and perhaps needs to be mentioned. The following news sources state that there are 32 spike mutations. SpookiePuppy (talk) 19:45, 25 November 2021 (UTC)

@PhiLiP: I've just had an afterthought, to ping the Stanford HIVDB team who created those superb genome map figures (https://en.wikipedia.org/wiki/Talk:Variants_of_SARS-CoV-2#Genome_maps) which showed the different SARS-CoV-2 variants, to make them aware of this discussion about the new variant (B.1.1.529). SpookiePuppy (talk) 20:14, 25 November 2021 (UTC)

@SpookiePuppy: Thanks! We are updating our site correspondingly. --PhiLiP (talk) 18:10, 28 November 2021 (UTC)

This reference might be useful, but the situation at the moment is so uncertain that I think it would be better to wait for a statement from the WHO. The effect of these mutations on neutralization needs to be tested. --Fernando Trebien (talk) 00:26, 26 November 2021 (UTC)

Six Southern African countries added to England's red list for travel; "100 cases confirmed through full sequencing of samples, up from just 10 reported on Wednesday, and signs of community transmission," 90% of cases in Gauteng. But still only VUM for WHO, not (yet) VOI or VOC. Boud (talk) 01:01, 26 November 2021 (UTC)

WHO will meet 26 Nov 2021 to decide if upgrading B.1.1.259 from VUM to VOI or VOC is justified.

There are already draconian travel restrictions all over the lot. Once WHO settles on the nomenclature, article drops. This thing came out of the gate fast...! https://www.theguardian.com/world/live/2021/nov/26/covid-news-live-new-variant-sparks-tougher-restrictions-in-india-and-singapore-ahead-of-who-meeting

kencf0618 (talk) 10:19, 26 November 2021 (UTC)

David Nabarro from the WHO: "it is going to be some weeks before we can say for certain whether our fears have any basis" [1] --Fernando Trebien (talk) 14:44, 26 November 2021 (UTC)

Maybe Responses to the COVID-19 pandemic in November 2021 is a good place to write about this subject for now. --Fernando Trebien (talk) 14:53, 26 November 2021 (UTC)

Certainly the UK's addition to their travel red list should be added to that page. Sideswipe9th (talk) 15:50, 26 November 2021 (UTC)

I've added a quick and dirty prècis focusing on Africa, but note that the impact is already global. It'll get its own article and timeline soon enough. (I refer to it as "Variant Nu" inasmuch as WHO hasn't officially announced the nomenclature.) kencf0618 (talk) 17:24, 26 November 2021 (UTC)

It's Omicron. https://www.bbc.com/news/live/world-59426277 kencf0618 (talk) 18:27, 26 November 2021 (UTC)

Proposal to add mutations to variant tables (total and in the spike protein)

Especially regarding the news for Variant B.1.1.529 with 32 mutations in the spike protein (for comparison Delta has 7+) also found in an Israeli Tourist returning from Malawi probably boostered with Pfizer–BioNTech it should be informative to add the info of known total and spike protein mutations to all variant comparison tables, so under Overview variants of concern (VOC) and Alerts for further monitoring (WHO) and maybe if possible also under Nomenclature SARS-CoV-2 corresponding nomenclatures. Understanding the range of mutational variability while remaining infectious of the SARS-CoV-2 virus is of utter importance combined with understanding how it effects natural and vaccine immunity in the variants of concern. --Krahax (talk) 08:54, 26 November 2021 (UTC)

The VOC table is a bit overloaded, but the Alerts table has plenty of space. However, being so atypical for B.1.1.529, I think it would be best to write this in prose for due weight. The number of mutations has theoretical implications - it "may" affect immunity from vaccines and the effectiveness of antibody therapies; but it also may not. Most mutations do not affect these characteristics, some do significantly, and the combined effect is difficult to predict. --Fernando Trebien (talk) 12:18, 26 November 2021 (UTC)

Thx for the response Ftrebien. AFAIK the immunity could be much more effected by the number of mutations in contrast to transmission-rate and severity which should depend more on the type of mutations. But of course any paper/expert review on this is welcome. I can't find the source but found an interesting 2D/3D plot of the mutation differences of Alpha, Beta, Gamma, Delta interesting. I'm not sure if it was in a paper with CC-BY(-SA) licence or in a media article. Not sure if possible but a comparison of the predicted post-Delta Scenarios would be informative to see if one model is close to the future reality. --Krahax (talk) 13:06, 26 November 2021 (UTC)

"Nu" Omicron variant

There are lots of mainstream news reports stating that B.1.1.529 has been named as the "nu" variant by the WHO. However, this seems to be jumping the gun; I see that the meeting to discuss this is happening right now (see [2]), and there is as yet no official word on this. As or when a "nu" variant is declared in WP:MEDRS, it should be added here. -- The Anome (talk) 17:03, 26 November 2021 (UTC)

And here we go: WHO has indeed declared B.1.1.529 a variant of interest, (see [3]) but assigned it the Greek letter omicron, not nu. Perhaps because "nu" sounds like "new"? -- The Anome (talk) 18:11, 26 November 2021 (UTC)

If they are skipping nu, the next should be xi not omicron. Agmartin (talk) 18:15, 26 November 2021 (UTC)

WHO published a statement about classifying B.1.1.529 as Omicron and a Variant of Concern. ― Aidan ⦿ (talk) 18:33, 26 November 2021 (UTC)

WHO recently named the new variant lineage #B11529 as "Omicron," but I have not been able to find here any info on the variant named "Xi", which is the preceding letter ib the Greek alphabet. 108.12.194.196 (talk) 19:00, 26 November 2021 (UTC)

Given that China's current leader is Xi Jinping, I think it's quite unlikely that the WHO will use the letter. I'm also guessing that it's likely that they skipped "nu" because it sounded too much like "new". -- The Anome (talk) 19:04, 26 November 2021 (UTC)

References

WHO skips nu and xi

WHO skips nu and xi.

https://www.rt.com/news/541479-who-nu-xi-omicron-covid/

https://nypost.com/2021/11/26/who-skips-two-letters-in-alphabet-in-naming-omicron-variant/ — Preceding unsigned comment added by 208.98.202.34 (talk) 05:46, 27 November 2021 (UTC)

yes your correct, rather odd on their part--Ozzie10aaaa (talk) 13:16, 28 November 2021 (UTC)

Hong Kong

In the article Hong Kong is said to have reported one case. There have actually been two, from the very moment it was announced that the B.1.1.529 strain was detected in their samples.[4][5][6][7] 1.64.47.144 (talk) 14:37, 27 November 2021 (UTC)