search

cov-lineages / pango-designation

Repository for suggesting new lineages that should be added to the current scheme
Other
1.05k stars 98 forks source link

Earliest date clarification #412

Closed truthcracy closed 2 years ago

truthcracy commented 2 years ago

It has been noted that the earliest date for this BA.2 is recorded as 2021-01-11
https://cov-lineages.org/lineage_list.html This has created some confusion as to the recent outbreak since januaryCould you please clarify what is this earliest date column relates to in the database.

corneliusroemer commented 2 years ago

Thanks for raising this. This is with almost certainty a classic hardcoded dating error, someone put 2021 instead of 2022.

truthcracy commented 2 years ago

Even the WHO variants page(https://www.who.int/en/activities/tracking-SARS-CoV-2-variants/) has the same error. last visit time UTC 2022-01-25 13:00 image

DataBeen commented 2 years ago

"Proposal to split B.1.1.529 to incorporate a newly characterised sibling lineage #361" includes sequences from November 2021 such as "South_Africa/CERI-KRISP-K032307/2021|EPI_ISL_6795834|2021-11-17", so perhaps the typo is in the month part of the date rather than in the year part, i.e. 2021-01-11 was intended to be input as 2021-11-11.

truthcracy commented 2 years ago

"Proposal to split B.1.1.529 to incorporate a newly characterised sibling lineage #361" includes sequences from November 2021 such as "South_Africa/CERI-KRISP-K032307/2021|EPI_ISL_6795834|2021-11-17", so perhaps the typo is in the month part of the date rather than in the year part, i.e. 2021-01-11 was intended to be input as 2021-11-11.

Well, it should be November

silcn commented 2 years ago

"Proposal to split B.1.1.529 to incorporate a newly characterised sibling lineage #361" includes sequences from November 2021 such as "South_Africa/CERI-KRISP-K032307/2021|EPI_ISL_6795834|2021-11-17", so perhaps the typo is in the month part of the date rather than in the year part, i.e. 2021-01-11 was intended to be input as 2021-11-11.

No, what's most likely happened is that a sequence from 2022-01-11 was input as 2021-01-11, and as a result it shows up as the "earliest sequence" on GISAID when it is not.

truthcracy commented 2 years ago

Even the WHO variants page(https://www.who.int/en/activities/tracking-SARS-CoV-2-variants/) has the same error. last visit time UTC 2022-01-25 13:00 image

who updated the site to fix the typo

corneliusroemer commented 2 years ago

@silcn is right, the earliest date is calculated automatically from GISAID date

lenaschimmel commented 2 years ago

I don't think it's worth a separate issue, but for BA.1 it even has the year 2020 🙃 Bildschirmfoto 2022-01-26 um 00 44 35

corneliusroemer commented 2 years ago

would be good to set a minimum for what's acceptable, or use the earliest designated sample date, or clock deviation to remove outliers 😬

rmcolq commented 2 years ago

I have updated the website with minimum date cutoffs for omicron. These had been set previously for B.1.1.529, but the same had not been applied to sublineages