How To: Create a Sankey diagram of variants circulating in your area
This thread will walk you through the process of using @Mike_Honey_'s visualization tool to generate a Sankey diagram for your country, province or State.🧵1/
An unrolled one-page web view for this long thread that may be easier to read or share can be found here ( ). 2/
Step 1: Go to Mike's variant visualization tool ( ). By default, Australia is chosen so you need to select your region of interest. 3/app.powerbi.com/view?r=eyJrIjo…
Step 2: Click on the V shape to the right of the current Continent, County, Location which by default is "Oceania + Australia" and you will see a drop-down list of continents and countries. Scroll down until you find your continent and then country of interest. 4/
Alternatively, type in the name of the country, province, or State that you are looking for in the search field to quickly narrow down the options. 5/
Step 3: In this example we are going to use the province of Ontario in Canada so you would click on the white box beside Ontario as shown in the image. 6/
After a few seconds the tool will load the sequences from your newly selected location, the default graph will be updated and now display the top 7 variant families for your location (see Date selector box for exact range). 7/
In this example the BA.2.86* family (the * means all variants of BA.2.86 which also includes any children of that variant such as JN.1) currently makes up about 56% of sequences in Ontario. 8/
Step 4: Switch to the Sankey diagram by clicking on the page number at the bottom of the web page as shown. 9/
Scroll down until you see "gisaid (nextclade) - sankey" in the page listing and click on that option as shown. 10/
It may take a minute to generate the Sankey diagram. By default all variants and their lineages during the time period selected are shown so may look pretty messy considering how many variants are circulating from unmitigated transmission and look something like this image. 11/
Step 5: Adjust the filters to make the Sankey diagram look how you want. The main filters you will want to adjust are the "Lineage Level" and "Minimum # Samples per Lineage to show" which can be found on the right side of the page. 12/
If you set the "Lineage Level" to 3 and 11 and the "Minimum # Samples per Lineage to show" to 60 it will make the Sankey diagram much less complex. 13/
The first few parent lineages on the left will be cut out along with a couple of child lineages on the right while only showing variants that have at least 60 samples that have been recorded. 14/
This may take a minute to process as it has to recalculate the entire diagram but you now have a less busy diagram. 15/
When the Lineage Level was 1, the Sankey showed BA.2 and XBB on the leftmost column, but now skips ahead two generations and starts with XBB.1.5, XBB.1.9, XBB.1.16, BA.2.75, and BA.2.86. 16/
Step 6: You will notice that most of the variants are labeled but some are not. By default the diagram only labels the most significant variants. If you want to see *all* labels, simply click on the "Show all labels" button in the bottom left corner as shown. 17/
Now all the labels will be shown. Sometimes even a filtered Sankey may seem a bit squished. 18/
You can maximize the Sankey diagram by hovering your mouse over the top right corner of the Sankey diagram, just under the year of the date and two icons will appear. Click on the box with the arrow pointing diagonally to the top right to maximize it. 19/
This will replace the right section of the visualization tool giving more space to draw the Sankey diagram. To return to the previous view click on "Back to report" in the top left corner of the diagram. 20/
Step 7: Enjoy your filtered Sankey diagram. 21/
Now that you have your filtered Sankey diagram, how do you interpret it? The height of each bar represents the relative difference in how many sequences there are for each variant in the time period that was selected. 22/
The longest bar in this example is XBB.1.9 "Hyperion" in the first column. Hovering over that bar shows there are 7,578 sequences found of XBB.1.9 including all of its descendants. 23/
While the original XBB.1.9 "Hyperion" variant itself is no longer circulating in Ontario, you can quickly see that the most common variants sequenced over the past two months in Ontario were based on the XBB.1.9 variant. 24/
Looking in column 4 you can see that most of those descendants come from the EG.5.1 "Eris" clan of variants. 25/
Remember that the Sankey diagram shows all the variants sequenced (except those filtered out) over the entire time period (the history of variants in the region) but does not identify the current most prevalent variant. 26/
To see the current most prevalent variants, you can use pages 1 and 2 of the tool using this guide ( ). 27/
The Sankey diagram also helps you trace the lineage of a variant. Changing the filter to set the max lineage level to 6 and the date range from Dec. 1, 2023 to present simplifies the diagram even further. 28/
If you look at the current most prevalent variant JN.1, you can trace its lineage back to BA.2.86.1 and then BA.2.86 to learn where this variant came from. 29/
Now that you know how to use @Mike_Honey_'s visualization tool, you can generate Sankey diagrams for almost anywhere in the world that still does COVID-19 genomic sequencing, which unfortunately has become less and less over time. 30/
Broad-Spectrum #COVID-19 #Therapeutics and #Prophylactics
An interesting presentation by Yunlong Richard Cao @yunlong_cao talking about their development of broadly #neutralizing #antibodies based on prediction of #viral #evolution ( ). 🧵1/
Read on to find out how their research team developed a neutralizing antibody solution (injection & nose spray) that works with all current variants, including the older SARS1 virus from 2003 and the currently globally dominant BA.2.86 Pirola clan of variants. 2/
An unrolled one-page web view for this long thread that may be easier to read or share can be found here ( ). 3/
#Wastewater levels in #Ottawa are extremely high and rising again with the JN.1 #Pirola variant now the most prevalent. #Hospitalization *for* #COVID-19 started declining just before the holidays but is increasing again.🧵1/
Our young children are especially vulnerable to respiratory virus infections. In Ottawa, despite children age 0-11 only making up 11% of the population, they accounted for 46% of respiratory related Emergency Department visits the week of Dec. 24 ( ). 2/ ottawapublichealth.ca/en/reports-res…
RSV levels in Ottawa wastewater were declining but now seeing another increase, but Influenza A levels increased quickly and are now twice the peak of last year already. Influenza B is now being detected in very small amounts. 3/
#Variant update for #Ontario, #Canada (to Dec. 26, 2023)
The BA.2.86.* #Pirola clan of variants is now the majority of sequences at 56%. JN.1 Pirola has taken 1st place at 38% with HV.1 "Eris" down to 18%. The next closest is JG.3.2 "Eris" at 7%. Graph tools by @Mike_Honey_🧵1/
Public Health Ontario got rid of their COVID data tool and replaced it with a respiratory virus tool making some of the previously reported data no longer available. 3/
Particulate Matter (PM) pollution event happening in Ontario and Quebec right now!
It seems PM levels have risen a lot overnight and whatever this event is stretches from southern Ontario (Toronto) to Ottawa to Montreal. Read more about the Model X and how I measured this below.
In case you missed them the first time, I'm looking back at what I thought were the most educational threads that I wrote over the past year starting with the oldest. 🧵1/
Why do people get (re)infected over and over again? COVID actively evades your immune system by suppressing parts of it to become invisible. Part 1 ( ) and Part 2 ( ). 2/
Outdoor Particulate Matter (#PM) Air Quality Monitoring
This thread explores the new Model X #outdoor PM air quality monitor made by #Canadian @IndianaTux at , what it does, costs, and examples showing the information it provides. #AQ #AirQuality 🧵1/ co2.click
An unrolled one-page web view for this long thread that may be easier to read or share can be found here ( ). 2/
After all the major wildfire smoke events this past summer, I wanted to get an air quality monitor that was suitable for outdoors so I would be better informed what the PM levels were in my neighbourhood. 3/