Variant spotters community announcement: new issue tracker for small, unclear, ambiguous, slow lineages

[FedeGueli]

https://github.com/sars-cov-2-variants/lineage-proposals/issues

To reduce the amount of work and to keep this page readable, we in the variant spotters community have set up via the impulse and the help of @corneliusroemer another page where everyone could transfer its smaller, less clear, ambigous , slower lineage proposals and open there the ones with not an immediate relevance due the fact they are small, unclear, ambiguous or slow to accumulate sequences (as the ones emerging in countries without a good sequencing intensity) or with no spike mutation or with mutations interesting cause rare but not immediately linked to a known beneficial effect.

There this kind of lineages could be discussed widely and also issues just to discuss are welcome.
The moderation will be the community itself with long time members being a direct link with the pango team to flag or ping interesting stuff.

It is nothing new , just instead of using one only repo for everything we will use this mai for proposal of evident interest or growth or numbers, while the less clear ones will be discussed by the community for a while and if they are recognized worth of being proposed here the community will push the author to officialize here her/his/ proposal.

TO BE CLEAR NO DESIGNATION COULD COME FROM THE OTHER REPO (https://github.com/sars-cov-2-variants/lineage-proposals/issues) the one only official Pango designation page will be this one.

We think that in this way also some chats with the newest members could be done without bearing on the pango committee ordinary ( or better said extraordinary work).

cc
@josetteshoenma @lenaschimmel @emilysmith @ilevade @AngieHinrichs @ryhisner @chrisruis @rmcolq @aineniamh
@shay671 @silcn @c19850727 @Sinickle @thomaspeacock @plomano @ciscorucinski @avizchl2 @krosa1910 @HynnSpylor @bitbyte2015 @rquiroga7 @RajLABN @karyakarte @banijolly @Simon-LoriereLab

[RajLABN]

Excellent idea!

[aviczhl2]

A great idea! I've always thinking about a place where we can discuss the following:

1 Collections of recombinants, sometimes two variants form a lot of recombinants with different breakpoints, these recombinants gather together but each of them only have a few seqs, this is a noteworthy situation but doesn't satisfy a proposal.

2 Usher misplacements of medium-sized clusters/already designated branches, also a noteworthy situation but doesn't satisfy a proposal.

3 Non-spike convergent evolutions(for example orf1a:L3829F or orf1a:T1788M ). Also a noteworthy situation but doesn't satisfy a proposal.

[krosa1910]

As this two-tiered system is supposed, I would like to further propose a standard of what could be left here in the repository.

I suggested that the lineage need to

1. Show a clear monophyletic evolutionary structure (if multiple lineages are proposed at one time, then for each lineage should be clear)
2. Have a considerable number of seqs ( >20?)
3. Have the latest presence relatively close ( being sequenced within 30 days?)
4. Have been spotted in different countries, (or at least some different regions within a country so it likely indicate spreading with long distance travel).

If some of us have an existing issue that does not satisfy this standard, I guess it is better to temporarily close it and move to the second tile repository, same for new issues. Possible exemption is given for recombination or in general very large saltation, say that having 10 nt mutation from reference basis variant.

[Memorablea]

What a great idea! I support this, thank you.

[FedeGueli]

Forgot @oobb45729 @BorisUitham @Outpfmance.

please if someone miss add here.

[FedeGueli]

As this two-tiered system is supposed, I would like to further propose a standard of what could be left here in the repository.

I suggested that the lineage need to

1. Show a clear monophyletic evolutionary structure (if multiple lineages are proposed at one time, then for each lineage should be clear)
2. Have a considerable number of seqs ( >20?)
3. Have the latest presence relatively close ( being sequenced within 30 days?)
4. Have been spotted in different countries, (or at least some different regions within a country so it likely indicate spreading with long distance travel).

If some of us have an existing issue that does not satisfy this standard, I guess it is better to temporarily close it and move to the second tile repository, same for new issues. Possible exemption is given for recombination or in general very large saltation, say that having 10 nt mutation from reference basis variant.

Yeah this is the standard but pls consider that BA.1.1.529 BA.4 BA.5 BA.2 75 BQ.1 XBB.1.5 XBB.2.3 and others have been caught with less than 20 Seqs,
If any of us has the sensation that something is worth a direct proposal keep this as main tool.
For everything else the list you have done is very good.

[FedeGueli]

If any of us has a closed issue that is worth a second chance (or discussion) please reopen it in the preproposal page.

[krosa1910]

Yeah this is the standard but pls consider that BA.1.1.529 BA.4 BA.5 BA.2 75 BQ.1 XBB.1.5 XBB.2.3 and others have been caught with less than 20 Seqs, If any of us has the sensation that something is worth a direct proposal keep this as main tool. For everything else the list you have done is very good.

I did not mean that one could not relocate back and forth here. If anyone found something that turned out to be successful later, they could create an issue in the second-tier repository at first, then track it until it fits the criterion to be observed here and move it here.

[FedeGueli]

cc @agamedilab @alurqu sorry i forgot to add you before.

[beansrowning]

Seems a fitting proposal. Was any thought given to just keeping everything on this repository and using GH Discussions instead?

https://github.com/features/discussions

That would also allow you to convert issues here into discussions if fitting, and vice-versa as proposals escalate. It seems less likely things would get lost that way, and probably maximize engagement.

There might have been other reasons to create a separate namespace and repository entirely for this purpose, but I thought to bring it up in case it wasn't considered.

[ciscorucinski]

Seems a fitting proposal. Was any thought given to just keeping everything on this repository and using GH Discussions instead?

@beansrowning Yes +1. I forgot about Discussions!

Since they already started using the new repo, I'm uncertain if switching again will be considered, but I think this is a potentially perfect fit. If you have any experience with using Discussions and have some protips, then please share them. Have you seen what it is like to move a discussion topic to an issue? I believe it would be preferable to have a fairly formal proposal in place before moving it to Issues (h/t @FedeGueli sans BA.2.75, BQ.1, XBB.1.5, XBB.2.3 like growth potential). Wondering about the process from a potential ad hoc discussion topic to a fairly formal proposal would look like.

(Added after the fact) I see that vercel/next.js Discussion have a RFC (Request for Comment) discussion structure which seems like a renamed Announcement category. This seems fit to flush out and update lineage proposals before they become final (move to Issues). However, only maintainers can create Announcement discussions.

How to Set Up Discussions

Enable

Within the Repo Settings Tab, you need to enable Discussions

Note: clicking on Set up Discussions doesn't do anything important. Just enable Discussions.

Customize

Now, you can go to the Discussion Tab and you can modify the categories, and add sections.

To edit: Click on the edit pencil icon next to the Categories text to edit the categories and sections.

Here I created 2 sections Community and Variant Trackers.

• For the Community Section, I moved all the categories to this section, except for Announcements.
• For the Variant Trackers Section, I have created a new category for regular discussion called Potential New Lineages. This is where we would find all the "small, unclear, ambiguous, slow lineages" discussions.

NOTE: I am testing out Discussions within my own repo (not this repo).

[FedeGueli]

@ciscorucinski thanks for your effort the other site is already running can you provide a link of your test version?

[corneliusroemer]

I'm aware of the Github discussions feature and consciously decided to go down the separate organization/repo route for a few reasons:

• People already know issues, issues can be linked well, discussions can be messy, the feature is not really that well developed (not a big fan of the feature)
• Opening discussions on this repo is a decision for the Pango committee to make, not for me - making a new unofficial issue tracker that is unassociated with Pango network can be done immediately
• In the past, Andrew locked issues when discussions went off-topic and/or became speculative, my feeling is that the preference is to not tie more liberal discussions to this repo
• It's easier to manage per repo notifications this way, one can decide to subscribe to all pango-designation issue comments but not to the pre-proposal ones in the other repo. If we add discussions, thinks could be more confusing.

[FedeGueli]

Thx @corneliusroemer . So we will keep that as official unofficial preproposal page.

[AngieHinrichs]

There is a web page with formally stated rules for lineage designation:

https://www.pango.network/the-pango-nomenclature-system/statement-of-nomenclature-rules/

They were originally written for the situation in 2020, and have had some small changes since then to adjust to the changing evolution of the virus, use of the UShER tree instead of a de novo tree for each lineage proposal, easier detection of recombinants when Delta and Omicron overlapped and so on. But if the written rules don't reflect the current practices, then they should be updated and/or our practices can be brought more back in line with stated rules.

[FedeGueli]

@AngieHinrichs i think that with the preproposal screening mostly we have solved the discrepancy between rules and proposals/discussions, lets see how it evolves ;)

[FedeGueli]

Hi everyone , i would like to thank everyone for how we are working together. 18 issues out of 175 preproposed ( the ones i briefly recognized ) have been designated starting from a preproposal issue,
We have less than 70 issues here in the official page and 90 in the other page all tracked and monitored.

Thanks @corneliusroemer for the idea and for your work.

[ryhisner]

And thank you, @FedeGueli, for all the work you've done in getting this new pre-proposal page started and operating it!

[FedeGueli]

Little update 35 lineages designated from the pre-proposal page out of 261 preproposals with 135 open issues and just 65 open issues on the main page (some of them stale) , I think ieverything is working very well. Thx @corneliusroemer and all the Pango team and all the variant spotters.

[LilKitt]

I am so glad you finally did this. Big improvement!

[Memorablea]

I think maybe we don't need to transfer to the homepage?

[FedeGueli]

I think maybe we don't need to transfer to the homepage?

If @corneliusroemer agrees i can add just a label "Proposed to Pango" or also just "Pango" on the other page with the same function of a proposal here. let me know what you think . @corneliusroemer @AngieHinrichs @thomasppeacock @InfrPopGen .

[Memorablea]

What I mean is that we don't need to transfer because it seems that the designated lineage is not specified according to the proposal?

There are many times when this is the case

[ciscorucinski]

Where would the designated label go? There are 2 labels currently used in conjunction with milestoning on cov-lineages/pango-designation and they are designated and proposed.

Edit: Ahh, scratch that. designated labels are in both repos

[FedeGueli]

Where would the designated label go? There are 2 labels currently used in conjunction with milestoning on cov-lineages/pango-designation and they are designated and proposed.

Edit: Ahh, scratch that. designated labels are in both repos

no no we were talking about the other repo. to me it is all ok. @Memorablea said we can jump from the other repo directly not to duplicate and mess things i think. to me it is ok both ways. One thing i am seeing that lineages when transferred here they seem a bit left behind , i sincerely when they are transferred here to me it's enough, ok done but i fear something important could be lost in tracking ( i have track of everything, but one people only could always mistake). So to me it is ok everything.

[DailyCovidCases]

Please close this issue

[DailyCovidCases]

Please close this issue