Covid Spread Simulation Tools

146 views
Skip to first unread message

Felizitas Eichner

unread,
Mar 26, 2020, 10:45:56 AM3/26/20
to virtual biohackathon COVID-19 2020
As my group is currently working on modelling the spread of infected and hospitalized patients in my area, I was wondering if we could just gather all modelling tools that we know about so far. As many other areas/hospitals are facing this task right now it would come very handy to already have tools on your hand and just figuring out the assumptions that most likely apply within your area.

As a second thought, one could also have a closer look at the tools and write something like reviews / pros & cons for each tool (however, many of them are updated almost daily so just a thought).

The ones I've came across so far in English Language:

Penn Medicine - COVID19 Hospital Impact Model for Epidemics

Gabriel Goh

Markus Schwehm and Martin Eichner together with the Landesgesundheitsamt Baden-Württemberg/Germany

Cheers
Feli

Pjotr Prins

unread,
Mar 26, 2020, 10:53:15 AM3/26/20
to Felizitas Eichner, virtual biohackathon COVID-19 2020
Sure thing. Add to group/topic on wiki!
> --
> You received this message because you are subscribed to the Google
> Groups "virtual biohackathon COVID-19 2020" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to [1]virtual-biohacka...@googlegroups.com.
> To view this discussion on the web, visit
> [2]https://groups.google.com/d/msgid/virtual-biohackathon/5a9409b4-760d
> -4e6d-9866-08e48ab903c7%40googlegroups.com.
>
> References
>
> 1. mailto:virtual-biohacka...@googlegroups.com
> 2. https://groups.google.com/d/msgid/virtual-biohackathon/5a9409b4-760d-4e6d...@googlegroups.com?utm_medium=email&utm_source=footer

Stephen Checkley

unread,
Mar 26, 2020, 11:13:45 AM3/26/20
to virtual biohackathon COVID-19 2020
Uri Alon has done some SIR modelling here:


In addition, Pedro Mendes of University of Connecticut is gathering papers for modelling at the moment, I think looking at assisting with logistics within the hospital:


There's also an R package available now for the data colated by Johns Hopkins:

Birgit Meldal

unread,
Mar 26, 2020, 11:32:21 AM3/26/20
to Stephen Checkley, virtual biohackathon COVID-19 2020

Hi Feli,

The UK health service (NHS) has just launched the Covid-19 symptom tracker: https://covid.joinzoe.com/

I have no idea what access anyone can have to this data. It only launched this week. I just signed up and apparently it asks you every day to say how you feel.

Birgit

--
You received this message because you are subscribed to the Google Groups "virtual biohackathon COVID-19 2020" group.
To unsubscribe from this group and stop receiving emails from it, send an email to virtual-biohacka...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/virtual-biohackathon/a1d4d843-23fb-411a-90b3-d6c3700fe54f%40googlegroups.com.
-- 
----------------------------------
Dr. Birgit Meldal
Senior Complex Portal & IntAct Curator
European Bioinformatics Institute (EMBL-EBI), European Molecular Biology Laboratory, Wellcome Genome Campus, Hinxton, Cambridgeshire CB10 1SD
United Kingdom

+44 1223 494107
bme...@ebi.ac.uk

http://www.ebi.ac.uk/intact/
http://www.ebi.ac.uk/complexportal

@complexportal
@intact_project
@bmeldal

Hilmar Lapp

unread,
Mar 26, 2020, 11:51:36 AM3/26/20
to Felizitas Eichner, virtual biohackathon COVID-19 2020
It’d also be nice to get hands on the model used by the NY Times.

  -hilmar

--
You received this message because you are subscribed to the Google Groups "virtual biohackathon COVID-19 2020" group.
To unsubscribe from this group and stop receiving emails from it, send an email to virtual-biohacka...@googlegroups.com.

Josiah Seaman

unread,
Mar 26, 2020, 12:50:33 PM3/26/20
to virtual biohackathon COVID-19 2020
I don't think this is relevant, but just in case: I'm the lead developer of the Animal Disease Spread Model (ADSM) developed at the USDA.   https://github.com/NAVADMC/ADSM/wiki/A-Quick-Start-Guide:-Running-the-sample-scenario  It likely has way more complexity than most of the numerical models being used with COVID-19.  However, Animal models tend not to be a good match for modeling human behavior, since a cow never gets on a plane to Japan by itself.  I could ask my epidemiologist colleagues what human disease model they recommend, but it's a complex field and I would assume everyone's already engaged.

To view this discussion on the web, visit https://groups.google.com/d/msgid/virtual-biohackathon/C0CCA0DB-839A-465A-A6CD-E194B040C9BE%40duke.edu.

Anna K

unread,
Mar 27, 2020, 6:19:39 AM3/27/20
to virtual biohackathon COVID-19 2020
I think this is a great plan!  Also, to think about: any model will need to be parametrized to the region of interest.  Getting data to parametrize a model can take time, as historical data are often not easily available or in a format that can easily be ingested into code.  For example, I've been compiling historical county data for Maryland (https://github.com/akonstodata/md_county_covid) using the COVID Tracking Project (https://covidtracking.com/), and even then, I have to manually update the numbers each since I am pulling them from the Dept. of Health MD website.

So, another possibility is, in conjunction with model review and testing, set up a group to compile and share historical data in formats that can be easily used for parameter estimation and deriving simple model spread statistics.

Best,
Anna

Hilmar Lapp

unread,
Mar 27, 2020, 5:40:32 PM3/27/20
to Anna K, virtual biohackathon COVID-19 2020
I love this, Anna. This is the kind of thing I was looking for for NC. Did you document the manual changes you had to apply to the COVID Tracking Project data?

I was considering an R package that wraps the COVID Tracking API. Would this be useful to have? Would anyone be interested in working on that (which could include testing, writing vignettes)?

  -hilmar

-- 
You received this message because you are subscribed to the Google Groups "virtual biohackathon COVID-19 2020" group.
To unsubscribe from this group and stop receiving emails from it, send an email to virtual-biohacka...@googlegroups.com.

Andrew Su

unread,
Mar 27, 2020, 7:13:49 PM3/27/20
to Hilmar Lapp, Anna K, virtual biohackathon COVID-19 2020
Hilmar,

I happened to be perusing around covidtracking.com (agreed, great resource!), and I saw this R package (https://github.com/aedobbyn/covid19us) listed on this page (https://covidtracking.com/software/)...

Best,
-a

To view this discussion on the web, visit https://groups.google.com/d/msgid/virtual-biohackathon/3EF861F3-7020-48B9-B8D1-C64E804645BE%40drycafe.net.

Anna K

unread,
Mar 27, 2020, 8:43:37 PM3/27/20
to virtual biohackathon COVID-19 2020
Thank you for your input Hilmar and Andrew!  I did indeed see the R package - and it's great if you want state-level data, but you can't drill deeper.  So, just depends on what you are looking for.  JHU also has data (world-wide and US based), but again may not have the area you are looking for (https://github.com/CSSEGISandData/COVID-19).  What I was interested was county level data - so I had to go to the individual screenshots in the history section for MD at covidtracking.com, and manually write down each day's numbers.  I don't know if this can be automated, but since this has been going for a relatively short duration, I thought it might be worthwhile to just take some time and convert screenshot data to csv files for help with parametrization if groups were interested in more local measures.  Another example are these state-by-state ICU projections provided by https://covid19.healthdata.org/projections.  For each state, they've compiled info on when/if a stay at home order occurred, educational facilities closed, non-essential services closed, etc.  So, getting that data (not sure whether it's directly available or would have to be copied down) would be a fairly quick effort, and again help contribute to parametrization.  That was kind of my gist - there's all this heterogeneous data floating around that, I think, when brought together, could be used to make the models that much more appropriate for the geographical location they are looking at.  

Anna

On Friday, March 27, 2020 at 7:13:49 PM UTC-4, Andrew Su wrote:
Hilmar,

I happened to be perusing around covidtracking.com (agreed, great resource!), and I saw this R package (https://github.com/aedobbyn/covid19us) listed on this page (https://covidtracking.com/software/)...

Best,
-a

On Fri, Mar 27, 2020 at 2:40 PM Hilmar Lapp <hl...@drycafe.net> wrote:
I love this, Anna. This is the kind of thing I was looking for for NC. Did you document the manual changes you had to apply to the COVID Tracking Project data?

I was considering an R package that wraps the COVID Tracking API. Would this be useful to have? Would anyone be interested in working on that (which could include testing, writing vignettes)?

  -hilmar

On Mar 27, 2020, at 6:19 AM, Anna K <konsto...@gmail.com> wrote:

I think this is a great plan!  Also, to think about: any model will need to be parametrized to the region of interest.  Getting data to parametrize a model can take time, as historical data are often not easily available or in a format that can easily be ingested into code.  For example, I've been compiling historical county data for Maryland (https://github.com/akonstodata/md_county_covid) using the COVID Tracking Project (https://covidtracking.com/), and even then, I have to manually update the numbers each since I am pulling them from the Dept. of Health MD website.

So, another possibility is, in conjunction with model review and testing, set up a group to compile and share historical data in formats that can be easily used for parameter estimation and deriving simple model spread statistics.

Best,
Anna



On Thursday, March 26, 2020 at 10:45:56 AM UTC-4, Felizitas Eichner wrote:
As my group is currently working on modelling the spread of infected and hospitalized patients in my area, I was wondering if we could just gather all modelling tools that we know about so far. As many other areas/hospitals are facing this task right now it would come very handy to already have tools on your hand and just figuring out the assumptions that most likely apply within your area.

As a second thought, one could also have a closer look at the tools and write something like reviews / pros & cons for each tool (however, many of them are updated almost daily so just a thought).

The ones I've came across so far in English Language:

Penn Medicine - COVID19 Hospital Impact Model for Epidemics

Gabriel Goh 

Markus Schwehm and Martin Eichner together with the Landesgesundheitsamt Baden-Württemberg/Germany

Cheers
Feli

-- 
You received this message because you are subscribed to the Google Groups "virtual biohackathon COVID-19 2020" group.
To unsubscribe from this group and stop receiving emails from it, send an email to virtual-biohackathon+unsub...@googlegroups.com.

-- 
Hilmar Lapp -:- lappland.io



--
You received this message because you are subscribed to the Google Groups "virtual biohackathon COVID-19 2020" group.
To unsubscribe from this group and stop receiving emails from it, send an email to virtual-biohackathon+unsub...@googlegroups.com.

Hilmar Lapp

unread,
Mar 27, 2020, 9:28:25 PM3/27/20
to Anna K, virtual biohackathon COVID-19 2020
I agree, county level data (or even more granular than that, but I haven’t seen those) are key for local decision making. I guess I need to dive more into the data that’s there.

There’s one newspaper in NC that tracks all cases, at a county level. It’s more up-to-date than the state’s DHHS. Source of infection is sometimes given, but not consistently. I’ve been curious how this looks for other states. 

 -hilmar

To unsubscribe from this group and stop receiving emails from it, send an email to virtual-biohacka...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/virtual-biohackathon/ae42723b-18ed-4905-8c7b-81698b704d81%40googlegroups.com.

Hilmar Lapp

unread,
Mar 27, 2020, 9:31:01 PM3/27/20
to Andrew Su, Anna K, virtual biohackathon COVID-19 2020


On Mar 27, 2020, at 7:10 PM, Andrew Su <a...@scripps.edu> wrote:


Wow, I missed it. It’s even on CRAN already. Kudos to the developer. That’s better than I was hoping for.

  -hilmar

Hilmar Lapp

unread,
Mar 28, 2020, 12:12:59 AM3/28/20
to Anna K, virtual biohackathon COVID-19 2020
The NY Times just released their data, including at the county level:

 -hilmar

Anna K

unread,
Mar 28, 2020, 11:09:06 AM3/28/20
to virtual biohackathon COVID-19 2020
What an amazing resource, thank you for sharing!

I think this thread contains a great compilation of data sources, btw.

I think, depending on what data you want, you might still need to dig into screenshots (e.g. Maryland just started to release age breakdowns of the cases) but overall the more resources we have available that allow easy data pulls, the more calibrated the models can become.

Best,
Anna


On Saturday, March 28, 2020 at 12:12:59 AM UTC-4, Hilmar Lapp wrote:
The NY Times just released their data, including at the county level:

 -hilmar
On Mar 27, 2020, at 9:28 PM, Hilmar Lapp <hl...@drycafe.net> wrote:

I agree, county level data (or even more granular than that, but I haven’t seen those) are key for local decision making. I guess I need to dive more into the data that’s there.

There’s one newspaper in NC that tracks all cases, at a county level. It’s more up-to-date than the state’s DHHS. Source of infection is sometimes given, but not consistently. I’ve been curious how this looks for other states. 

 -hilmar

To unsubscribe from this group and stop receiving emails from it, send an email to virtual-biohackathon+unsub...@googlegroups.com.

-- 
Hilmar Lapp -:- lappland.io



v

unread,
Mar 28, 2020, 11:59:01 AM3/28/20
to Hilmar Lapp, Anna K, virtual biohackathon COVID-19 2020
It looks like there is a python module too, although I didn't see it listed on the site. https://pypi.org/project/covidtracking/. It might be a temporary bloop with GitLab, but I wasn't able to see the code: https://gitlab.com/blfordham/covid-tracking

image.png


Noushin Nabavi

unread,
Mar 28, 2020, 4:58:24 PM3/28/20
to virtual biohackathon COVID-19 2020
Does anyone know if there's someone out there who's worked on an R programming solution for simulation/modelling?
I ask because loading, tidying, visualizing covid data in R is there and modelling is missing.

LJ.Garcia

unread,
Mar 28, 2020, 6:40:39 PM3/28/20
to Anna K, virtual biohackathon COVID-19 2020
Hi there, 

Not sure if you had already but in case you have not, could you please add an entry to the wiki with the compilation of resources from this thread? It sound like a good entry to have at hand for everyone. 

Thanks! 
 -hilmar

  -hilmar

To unsubscribe from this group and stop receiving emails from it, send an email to virtual-biohackathon+unsubscr...@googlegroups.com.

-- 
Hilmar Lapp -:- lappland.io




--
You received this message because you are subscribed to the Google Groups "virtual biohackathon COVID-19 2020" group.
To unsubscribe from this group and stop receiving emails from it, send an email to virtual-biohackathon+unsubscrib...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "virtual biohackathon COVID-19 2020" group.
To unsubscribe from this group and stop receiving emails from it, send an email to virtual-biohackathon+unsubscrib...@googlegroups.com.

-- 
Hilmar Lapp -:- lappland.io




-- 
Hilmar Lapp -:- lappland.io



--
You received this message because you are subscribed to the Google Groups "virtual biohackathon COVID-19 2020" group.
To unsubscribe from this group and stop receiving emails from it, send an email to virtual-biohackathon+unsub...@googlegroups.com.

Felizitas Eichner

unread,
Mar 29, 2020, 10:48:52 AM3/29/20
to virtual biohackathon COVID-19 2020
Hi all,

I saw that most of the data/tools/packages from this thread have already been added to the biostats wiki. I think that is the right spot for this and I just added the resources that were missing there. Please feel free to add further resources in case I missed something. We could of course also open a new entry in the wiki, but I have the feeling there already quite a lot. Maybe one can gather within biostats if interested in the overarching topic of data preparation/extraction, SIR modelling ect. and then surely subgroups will arise from there.

Cheers,
Feli


Am Samstag, 28. März 2020 23:40:39 UTC+1 schrieb LJ.Garcia:
Hi there, 

Not sure if you had already but in case you have not, could you please add an entry to the wiki with the compilation of resources from this thread? It sound like a good entry to have at hand for everyone. 

Thanks! 
 -hilmar

  -hilmar

To unsubscribe from this group and stop receiving emails from it, send an email to virtual-biohackathon+unsub...@googlegroups.com.

-- 
Hilmar Lapp -:- lappland.io




--
You received this message because you are subscribed to the Google Groups "virtual biohackathon COVID-19 2020" group.
To unsubscribe from this group and stop receiving emails from it, send an email to virtual-biohackathon+unsub...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "virtual biohackathon COVID-19 2020" group.
To unsubscribe from this group and stop receiving emails from it, send an email to virtual-biohackathon+unsub...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/virtual-biohackathon/ae42723b-18ed-4905-8c7b-81698b704d81%40googlegroups.com.

-- 
Hilmar Lapp -:- lappland.io




-- 
Hilmar Lapp -:- lappland.io



Felizitas Eichner

unread,
Mar 29, 2020, 10:54:17 AM3/29/20
to virtual biohackathon COVID-19 2020
And also a wiki entry for "Resources" has been added.

BW
Feli
Reply all
Reply to author
Forward
0 new messages