Defining transmission intensity

Implementation in MultiMalModPy

🔍 For a detailed overview of how each model handles transmission intensity, refer to the Transmission Intensity description under malaria models.

In MultiMalModPy, the user can specify the level of transmission via the following two parameters in setup_sim.py: - target_output_name – Defines the epidemiological metric used for calibration.
- target_output_values – Specifies the target values for the selected metric.

Example: 

target_output_name = 'prevalence_2to10'
target_output_values = [0.25]

In this example, the simulation is calibrated to achieve a 2-to-10-year-old malaria prevalence of 25% (0.25).

Matching defined transmission output targets

The target_calibration directory includes interpolation_data.csv for each of the models included in the framework. This file is generated during the transmission calibration process when running: 

python launch_sim.py -r calibrun

The table below provides an example of calibration results for EMOD, showing key epidemiological metrics related to malaria transmission.

input_target seasonality cm_clinical simulatedEIR prevalence_2to10 prevalence severe_incidence n_total_mos_pop clinical_incidence clinical_incidence_U5 modelname pop_size importation
8.44553654 perennial 0.45 0.343540425 0.020460788 0.044149221 0.000403342 56526.55781 0.265060585 0.088925115 EMOD 50000 FALSE
10.9762974 perennial 0.45 0.932994042 0.05381089 0.106625617 0.000708954 73407.88671 0.636199389 0.235184127 EMOD 50000 FALSE
14.2654175 perennial 0.45 1.710211503 0.095421294 0.173174509 0.000903154 95345.36855 0.956651654 0.423187833 EMOD 50000 FALSE
18.5401442 perennial 0.45 2.573332919 0.138401713 0.224445658 0.001041937 123878.2838 1.113905394 0.624657857 EMOD 50000 FALSE
24.095821 perennial 0.45 3.623527973 0.185692034 0.266564637 0001239512 160984.3319 1.22631187 0.853578362 EMOD 50000 FALSE
31.316293 perennial 0.45 5.016965242 0.242720614 0.307751194 0.001423436 209179.4093 1.326620187 1.145714057 EMOD 50000 FALSE
40.7004273 perennial 0.45 6.53350783 0.295704873 0.338061961 0.001647561 271863.5711 1.387121162 1.429262657 EMOD 50000 FALSE

Column Descriptions

  • input_target: The transmission target input value for the simulation.
  • seasonality: The seasonality type, indicating the transmission pattern (e.g., perennial).
  • cm_clinical: The clinical case management rate.
  • simulatedEIR: The estimated entomological inoculation rate based on the simulation.
  • prevalence_2to10: Prevalence of malaria in children aged 2 to 10 years.
  • prevalence: Overall prevalence of malaria in the population.
  • severe_incidence: The incidence rate of severe malaria cases.
  • n_total_mos_pop: Total number of mosquitoes in the population.
  • clinical_incidence: Overall clinical incidence rate of malaria.
  • clinical_incidence_U5: Clinical incidence rate of malaria in children under 5 years old.
  • modelname: The name of the model used in the simulation (e.g., EMOD).
  • pop_size: The size of the population used in the simulation.
  • importation: Indicates whether importation of cases is considered (TRUE or FALSE).

Defining your own transmission target

If you need to run simulations beyond the predefined datasets, follow this process: 

python launch_sim.py -r calibrun

By default, this runs all three models with the settings specified in setup_sim.py.
Key parameters to configure include:
- Population size (pop_size)
- Burn-in duration - Seasonal profiles - Case management levels

Once the simulations complete, new .csv files will be generated in the target_calibration folder for each model. These files are named after your most recent calibration run.

You should review and decide whether to integrate them into your interpolation_data.csv files.

Special Considerations for OpenMala

📝 OpenMalaria Consideration: It is recommended to discard OpenMalaria simulation outputs where the annual *Pf*PR (_Plasmodium falciparum_ prevalence in ages 2-10) exceeds 0.7, as these can not be reliably simulated using OpenMalaria.
📝 EMOD Consideration 1: If your calibration simulations for EMOD did not include importation, check whether any lower xtemp simulations resulted in malaria elimination. To do this, open the relevant simulation_outputs/EMOD/mmmpy_yr.csv file and look for cases where simulatedEIR equals 0, indicating elimination.
📝 EMOD Consideration 2: If even one of the simulations you ran resulted in elimination, remove that xtemp value from interpolation_data.csv, as it cannot reliably sustain malaria transmission. If the lowest level of malaria EMOD can sustain does not meet your expectations, consider adjusting the max_larval_capacity values in the habitat_setup function (EMOD/functions/EMOD_functions.py).

Finalizing and Running Production Simulations

Once you have finalized your interpolation_data.csv files, you can proceed with running production simulations.