Omicron in Israel in 2020

[yuval-harpaz]

Displaying only recent variants I was surprised to see a bump in Oct 2020. Is this a bug? Or is it an early version of Omicron somehow tracked by the system? It seems to come from this json, look up Israel, 12 Oct 2020 (1340 cases)

[emmahodcroft]

Hi @yuval-harpaz Thanks for pointing this out! This is definitely something weird happening with the data - we have catches in place so that data from Variants that is dated from before they existed shouldn't get through. I'm not sure why this is! However, I've found the offending sequences and have manually excluded them. I'm running an update now and hopefully by the end of the day these odd Omicrons sequences won't be showing up anymore!

[yuval-harpaz]

Hi, after the update, the culprit is still there. Please see the following segment in perCountryDataCaseCounts.json }, { "percent_total_cases": 0.0002140983782047851, "stand_estimated_cases": { "20A.EU2": 0, "20A/S:439K": 0, "20A/S:98F": 0, "20B/S:732A": 0, "20E (EU1)": 0, "20H (Beta, V2)": 0, "20I (Alpha, V1)": 0, "20J (Gamma, V3)": 0, "21A (Delta)": 0, "21B (Kappa)": 0, "21C (Epsilon)": 0, "21D (Eta)": 0, "21G (Lambda)": 0, "21H (Mu)": 0, "21I (Delta)": 0, "21J (Delta)": 0, "21K (Omicron)": 0, "21L (Omicron)": 0, "22A (Omicron)": 1005, "22B (Omicron)": 0, "22C (Omicron)": 502, "others": 503 }, "stand_total_cases": 2010, "total_sequences": 4, "week": "2020-10-12" },

[emmahodcroft]

Hi @yuval-harpaz - Apologies for the confusion. I did one update this morning, but it didn't include this fix! A new run (which includes this update) is running now, and hopefully will be up in a few hours!

[emmahodcroft]

Hi @yuval-harpaz - just to let you know, the new update is now live, and the offending sequences should be gone! Thanks again for alerting me to it!

[yuval-harpaz]

Perfect, thanks. Here is another thanks on twitter.