djgroen / facs Goto Github PK

Mutations currently support changes in infection rate, but should contain complete disease definitions.

The infection function should weight between mutations, and select the specific disease properties depending on the disease being spread.

mutations/ strains do not have to remain identical between infection events, as that could add extra bias to the code (we seldom know where mutations are exactly introduced)

Currently hard coded.

Currently, households have members with a completely random age. This can lead to households with only children for example.

We should introduce a method to generate more realistic households.

My suggestion would be to make the first member 18+ by default, and then to change the probability of child / adult generation for the other household members to ensure that the distributed in demographics.csv is fully adhered to.

To do this, we need to probably split the eco.ages structure into two distributions, one for ages 0-17 and one for ages 18+.

Hi Derek,

Here is the output that I get.

fabsim localhost facs_locationplot:klaipeda_localhost_1
[loading plugin] FabDummy ...
[loading plugin] FabCovid19 ...
Error: failed to import settings module ((<class 'SyntaxError'>, SyntaxError('invalid syntax', ('/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/plugins/FabCovid19/VVP/facs_VVP_ensemble_validation.py', 143, 22, ' facs_vvp_sif,\n')), <traceback object at 0x7f59956ff0c0>))
The FabCovid19 ensemble VVP functionalities are not imported as a result.
[Executing task] facs_locationplot
calling task facs_locationplot from plugin FabCovid19
╭───── New/Updated environment variables from FabCovid19 plugin ─────╮
│ env : │
│ --- facs_location : /home/arindam/Dropbox/Brunel/FACS/facs/facs/ │
│ +++ facs_location : /home/arindam/Dropbox/Brunel/FACS/facs │
│ env : │
│ +++ facs_args is a new added key │
│ env : │
│ +++ facs_validation is a new added key │
╰────────────────────────────────────────────────────────────────────╯
calling task locate_results_dir from plugin FabCovid19
╭─ New/Updated environment variables from FabCovid19 plugin ─╮
│ │
╰────────────────────────────────────────────────────────────╯
/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1 ['RUNS'] ['klaipeda-extend-lockdown-1.html']
/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS ['extend-lockdown-1-0.475_1', 'extend-lockdown-1-0.475_2', 'extend-lockdown-1-0.475_4', 'extend-lockdown-1-0.475_3', 'extend-lockdown-1-0.475_5'] []
/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475_1 ['covid_data'] ['covid_out_deaths_0.csv', 'env.log', 'covid_out_infections_0.csv', 'covid_out_recoveries_0.csv', 'out.csv', 'offices.csv', 'simsetting.csv', 'env.yml', 'covid_out_hospitalisations_0.csv', 'klaipeda_localhost_1_extend-lockdown-1-0.475_1.sh']
file found at: /home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475_1/covid_out_infections_0.csv
/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/fabsim/base/decorators.py:127: DtypeWarning: Columns (0,1,2) have mixed types.Specify dtype option on import or set low_memory=False.
return func(*args, **kwargs)
/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475_1/covid_data [] ['needs.csv', 'measures_uk.yml', 'age-distr.csv', 'vaccinations_example.yml', 'test_cases.csv', 'disease_covid19.yml', 'building_types_map.yml', 'klaipeda_buildings.csv']
/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475_2 ['covid_data'] ['covid_out_deaths_0.csv', 'env.log', 'covid_out_infections_0.csv', 'covid_out_recoveries_0.csv', 'out.csv', 'offices.csv', 'simsetting.csv', 'env.yml', 'covid_out_hospitalisations_0.csv', 'klaipeda_localhost_1_extend-lockdown-1-0.475_2.sh']
file found at: /home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475_2/covid_out_infections_0.csv
/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475_2/covid_data [] ['needs.csv', 'measures_uk.yml', 'age-distr.csv', 'vaccinations_example.yml', 'test_cases.csv', 'disease_covid19.yml', 'building_types_map.yml', 'klaipeda_buildings.csv']
/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475_4 ['covid_data'] ['covid_out_deaths_0.csv', 'env.log', 'covid_out_infections_0.csv', 'covid_out_recoveries_0.csv', 'out.csv', 'offices.csv', 'simsetting.csv', 'klaipeda_localhost_1_extend-lockdown-1-0.475_4.sh', 'env.yml', 'covid_out_hospitalisations_0.csv']
file found at: /home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475_4/covid_out_infections_0.csv
/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475_4/covid_data [] ['needs.csv', 'measures_uk.yml', 'age-distr.csv', 'vaccinations_example.yml', 'test_cases.csv', 'disease_covid19.yml', 'building_types_map.yml', 'klaipeda_buildings.csv']
/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475_3 ['covid_data'] ['covid_out_deaths_0.csv', 'env.log', 'covid_out_infections_0.csv', 'covid_out_recoveries_0.csv', 'out.csv', 'offices.csv', 'simsetting.csv', 'env.yml', 'covid_out_hospitalisations_0.csv', 'klaipeda_localhost_1_extend-lockdown-1-0.475_3.sh']
file found at: /home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475_3/covid_out_infections_0.csv
/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475_3/covid_data [] ['needs.csv', 'measures_uk.yml', 'age-distr.csv', 'vaccinations_example.yml', 'test_cases.csv', 'disease_covid19.yml', 'building_types_map.yml', 'klaipeda_buildings.csv']
/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475_5 ['covid_data'] ['covid_out_deaths_0.csv', 'env.log', 'covid_out_infections_0.csv', 'covid_out_recoveries_0.csv', 'out.csv', 'offices.csv', 'simsetting.csv', 'env.yml', 'covid_out_hospitalisations_0.csv', 'klaipeda_localhost_1_extend-lockdown-1-0.475_5.sh']
file found at: /home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475_5/covid_out_infections_0.csv
/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475_5/covid_data [] ['needs.csv', 'measures_uk.yml', 'age-distr.csv', 'vaccinations_example.yml', 'test_cases.csv', 'disease_covid19.yml', 'building_types_map.yml', 'klaipeda_buildings.csv']
Traceback (most recent call last):
File "/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/fabsim/bin/fabsim", line 46, in
sys.exit(fabsim_main.main())
File "/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/fabsim/base/fabsim_main.py", line 150, in main
env.exec_func(*env.task_args, **env.task_kwargs)
File "/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/fabsim/base/decorators.py", line 74, in wrapper
return func(*args, **kwargs)
File "/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/fabsim/base/decorators.py", line 127, in wrapper
return func(*args, **kwargs)
File "/home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/plugins/FabCovid19/visualisation/PlotByLocationType.py", line 63, in facs_locationplot
for i in range(pdf['time'].min(), pdf['time'].max()):
File "/home/arindam/.local/lib/python3.8/site-packages/pandas/core/generic.py", line 10827, in min
return NDFrame.min(self, axis, skipna, level, numeric_only, **kwargs)
File "/home/arindam/.local/lib/python3.8/site-packages/pandas/core/generic.py", line 10350, in min
return self._stat_function(
File "/home/arindam/.local/lib/python3.8/site-packages/pandas/core/generic.py", line 10345, in _stat_function
return self._reduce(
File "/home/arindam/.local/lib/python3.8/site-packages/pandas/core/series.py", line 4391, in _reduce
return op(delegate, skipna=skipna, **kwds)
File "/home/arindam/.local/lib/python3.8/site-packages/pandas/core/nanops.py", line 155, in f
result = alt(values, axis=axis, skipna=skipna, **kwds)
File "/home/arindam/.local/lib/python3.8/site-packages/pandas/core/nanops.py", line 410, in new_func
result = func(values, axis=axis, skipna=skipna, mask=mask, **kwargs)
File "/home/arindam/.local/lib/python3.8/site-packages/pandas/core/nanops.py", line 1017, in reduction
result = getattr(values, meth)(axis)
File "/home/arindam/.local/lib/python3.8/site-packages/numpy/core/_methods.py", line 44, in _amin
return umr_minimum(a, axis, None, out, keepdims, initial, where)
TypeError: '<=' not supported between instances of 'int' and 'str'

This is useful for vis and analysis purposes :).

When household choose parks, shops etc., size should also play a partial role, in addition to distance.

@mzrghorbani I figured this could be a good issue to describe what you are adding or changing to FACS to support measles modelling? We can spin off separate issues here if you need support on the implementation of specific features (or selective disablement of other ones) :).

I am not attaching this to the FACS 2.0 milestone for now.

The individual deaths and recoveries of agents needs to be included in the simulation outputs

• Add 2-4 extra simulations to the autovalidator.
• Check if validation data files are present.
• Do a test run using fabsim archer2 validate_facs:cores=24 and see what the error is.

The additional simulation configurations should be added here: https://github.com/djgroen/FabCovid19/tree/master/config_files/facs_validation/SWEEP

Expose the following constants as runtime parameters in run.py and run_grid.py:

read_building_csv.read_building_csv(e, building_file, "{}/building_types_map.yml".format(data_dir), house_ratio=house_ratio, workspace=20, office_size=2500, household_size=household_size, work_participation_rate=0.5)

Not necessary for STAMINA trial, but desirable on the medium term.

It seems that hospitals are chosen randomly based on only size and is not influenced by distance. Is this intended?

In this issue, I am going to list all the things we would like to update in FACS this time around:

• Develop new scenarios for opening up, or local lockdown.
• Update existing PHI's:
  • Partial school reopening on June 1st (50% primary school)
  • Test and tracing effective June 1st, ramped up to 130k by end of the month (https://www.gov.uk/government/news/nhs-test-and-trace-contacts-more-than-130000-people) 1b57164
  • June 15th, secondary schools offer face to face contact (https://schoolsweek.co.uk/schools-will-reopen-on-june-1-prime-minister-confirms/)
  • June 15th, face masks compulsory on public transport
  • June 15th, non-essential retail reopens
  • July 1st, flexible furlough (some workers brought back)
  • July 4th, pubs, restaurants and hairdressers reopen.
  • July 6th, quarantine restrictions lifted for many foreign countries.
  • Outdoor gyms and playgrounds open July 11th
  • Face masks compulsory in shops as of July 14th
  • Beauty salons etc. open July 13th
  • Indoor gyms and pools open July 25th
• Incorporating latest findings scientifically on airborne transmission and masks.
  • airborne transmission: https://www.tandfonline.com/doi/full/10.1080/22221751.2020.1777906
  • masks: https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)31142-9/fulltext 1b57164
  • reduced odds for outside transmission: https://www.medrxiv.org/content/10.1101/2020.02.28.20029272v2 5ac131b
• Updating mortality rates, see #2 d55c2b9
• Developing a better model for social distance compliance?
• Output individual deaths and recoveries #5 efc19f3
• Develop explicit model for interactions in pubs, restaurants, and inter-household indoor visits (the first two are currently implicit in leisure, and the last one is not in the model yet).
• Added more realistic transport multipliers for June/July and after (it now goes up from 10% to 20%). 7ac4469

This is to enable model execution with start point after the initial outbreak of the pandemic as a whole.

Perhaps the parameters need some tweaking but currently the visit probability in atleast some locations is more than one. This is caught by the error handler here:

which traces back to

Should I set it to 1 if it is more, or should we change the visit times?

Currently, persons go to the nearest medical facility when either visiting or being hospitalised. Sometimes these facilities are rather small, creating asymmetric effects.

My aim is to implement the following:

• When at home: random chance to visit medical facility irrespective of type (some will visit hospitals, some clinics).
• When hospitalised: exclude non-hospital medical facilities from selection.

Imran, what to do you think of this approach?

New mutant variants may have levels of immune evasiveness. Some people immune to one variant may not be immune to other variants.

check that vac_no_symptoms and vac_no_transmission are in range 0.0--1.0.

• measures_uk.yml to measures.yml
• disease_covid19.yml to disease.yml
• vaccinations_example.yml to vaccinations.yml

As a proof of concept, I tried to multiply each of the school areas by a thousand in the region of Kent (and named it test_kent) in FabCovid19 repo. The results did not change qualitatively. This is similar to the result reported by Tasin (@1702609) on Teams when he was working with Turkey.

What could be causing this issue?

https://www.thelancet.com/journals/laninf/article/PIIS1473-3099(20)30833-1/fulltext
https://aip.scitation.org/doi/10.1063/5.0026360

When the starting_infections argument has a leading 0, we should read it as a float and use it as a ratio.

As per Maziar's suggestion, we should split off the source code a bit such that facs.py is a bit less overwhelming in size, and the code is easier to navigate for people other than me ;).

• Revise microscale model equations based on https://airborne.cam/ https://royalsocietypublishing.org/doi/10.1098/rspa.2020.0584 https://science.sciencemag.org/content/early/2020/10/02/science.abf0521
• Check for possible optimisations to reduce time to completion (e.g. aggregating RNG calls)
• Deploy code on new resources
• Design an approach to incorporate new strains / mutations explicitly, and reduction of vaccine efficacy.
• test vaccine response ranges.

Add support for care homes as a location type in FACS.

This is a request to introduce new infections in a specific location or area in FACS.

@arindamsaha1507 before I proceed with designing and implementing this, could you give me a few ideas on how you would like to define this in a run.

E.g. would you like an input Yaml file in a certain format for this? Feel free to propose something :).

Hospitalised people should not propagate infections across their household as much as non-hospitalised people.

Can the blood type be an agent parameter?

https://hms.harvard.edu/news/covid-19-blood-type

• change nearest visiting location routine, to do school by classes. (Derek)
• change household size to 8, with reduced interaction frequencies. (Imran)
• include accounting for outside visitors. (Imran, optional)
• revise needs to include lecture / lab/ library? (Imran)
• introduce teaching space concept. (Imran)
• build in summer closure
• doube-check and fix consistency of lockdown measures in campus setting.

Continuation of issue #34

Go through the course and remove unused functions, reduce coupling etc.

Integration of Measles into FACS is a significant step towards enhancing our disease modeling capabilities. This integration will allow us to leverage Measles disease parameters and expand the scope of our simulations.

Goals and Objectives:

1. Improve disease modeling in FACS.
2. Expand our simulation capabilities.
3. Enhance the accuracy of our results.
4. Improved simulation validation.

Key Steps:

1. Review disease parameters in Measles.
2. Identify relevant parameters for integration.
3. Design the integration process.
4. Implement code changes.
5. Thoroughly test and validate the integrated solution.
6. Enhance the FACS validator to accommodate new parameters.

Timeline:

• We'll aim to complete the initial review and parameter identification within the next two weeks.
• Detailed design and coding phases will follow, with regular updates in this thread.

Responsibility:

I'll be overseeing this integration, and I'm open to collaboration. If you have specific expertise or insights to contribute, please feel free to join in.

Feedback and Collaboration:

Your feedback, suggestions, and concerns are invaluable. Let's keep the communication flowing in this thread as we make progress.

Hi Derek,

I have been getting this bug while running the code via FabSim:

RuntimeError:
local() encountered an error (return code 2)while executing 'ssh -Y -p 22 arindam@localhost '/bin/true || true && /home/arindam/Dropbox/Brunel/FACS/FabSim/FabSim3/localhost_exe/FabSim/results/klaipeda_localhost_1/RUNS/extend-lockdown-1-0.475/klaipeda_localhost_1_extend-lockdown-1-0.475.sh''

The exact command I ran was

fabsim localhost covid19_ensemble:cheshire_east,TS='extend-lockdown',TM='1',cores=1,replicas=1,ci_multipliers=0.3,simulation_period=100

I see this bug since the update to the code about 4 days ago.

The above-mentioned line results in the covid_out files to be written in the current directory. Would it not be better if they are written in the results directory?

Dear Dr Derek Groen,
I am Shakir Bilal working at University of South Florida, In Tampa. Our group has been working on a C++ based ABM simulator and seeks to perform sensistvty analysis using the facs tool that you have developed. I have been working with Dr Arindam Saha while he was here at USF last year.

The current issue is with installing the facs on my machine with python 3.8.10 on Ubuntu. The installation process is stuck with the following two issues:
`

allgatherv.c: In function ‘MPI_Allgatherv’:
allgatherv.c:50:1: internal compiler error: in init_alias_analysis, at alias.c:3468
50 | }
| ^

allgatherv.c:50:1: internal compiler error: Segmentation fault
0x7fba204e408f ???
/build/glibc-wuryBv/glibc-2.31/signal/../sysdeps/unix/sysv/linux/x86_64/sigaction.c:0
gcc: internal compiler error: Segmentation fault signal terminated program cc1
Please submit a full bug report,
with preprocessed source if appropriate.
See file:///usr/share/doc/gcc-9/README.Bugs for instructions.
make[2]: *** [Makefile:1876: allgatherv.lo] Error 1
make[2]: Leaving directory '/home/shakir/Downloads/openmpi-4.1.2/ompi/contrib/libompitrace'
make[1]: *** [Makefile:3555: all-recursive] Error 1
make[1]: Leaving directory '/home/shakir/Downloads/openmpi-4.1.2/ompi'
make: *** [Makefile:1901: all-recursive] Error 1
Collecting mpi4py
Using cached mpi4py-3.1.5.tar.gz (2.5 MB)
Installing build dependencies ... done
Getting requirements to build wheel ... done
Preparing metadata (pyproject.toml) ... done
Building wheels for collected packages: mpi4py
Building wheel for mpi4py (pyproject.toml) ... error
error: subprocess-exited-with-error

× Building wheel for mpi4py (pyproject.toml) did not run successfully.
│ exit code: 1
╰─> [242 lines of output]
'
2)
'
#include "Python.h"
^~~~~~~~~~
compilation terminated.
error: command '/home/shakir/sfw/MPICH/bin/mpicc' failed with exit code 1
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
ERROR: Failed building wheel for mpi4py
Failed to build mpi4py
ERROR: Could not build wheels for mpi4py, which is required to install pyproject.toml-based projects'

• Add vaccine_effect_time from YML as attribute in FACS.
• Move hard-coded mutations to YML file.
• Revise hospitalisation protection factor.
• (add support for boosters)
• Put vacc duration and immunity duration in YML files.

They are long overdue...

Suggestions are welcome here.