NGL, it wasn't much data anymore with how little monitoring is happening in florida, but it's still just about all that was reliable.

https://twitter.com/BiobotAnalytics/status/1791944840789291116

People are leaving replies asking them to reconsider. Hopefully they do.

  • sovietknuckles [they/them]
    ·
    4 months ago

    The CDC's Wastewater Surveillance dashboard can answer simple questions like whether whether line go up nationally or in specific counties, but some graphs don't have direct replacements. The year-over-year dashboard could not predict future spikes, but it was helpful predicting rate of decline:

    Show Year-over-year chart

    That one can be remade using existing data, though it lacks variant data.

    The one that I paid more attention to was the COVID-19 Variant Trends over Time chart, which illustrates transmission advantage of different variants by how quickly they become dominant:

    Show COVID variants dashboard

    CoVariants.org maybe has a comparable chart (It doesn't show KP.2 yet, but neither did Biobot). Unlike Biobot, they have variant charts for other countries and individual US states.

    Show CoVariants.org US chart

    The CDC's 2-week variant chart does include KP.2/KP.3, but because it doesn't group sublineages like Biobot did, it might not not be as coherent over a larger period than 2 weeks.

    Show CDC 2-week variant chart