mpox-model

This repository contains Python code for the study "Modeling the 2022 Mpox Outbreak with a Mechanistic Network Model".

For any further inquiries, please email [email protected].

File Overview

Naming Conventions

All the files follow the same file naming convention following "mpox_": behavior change start time, behavior change scenario, reduction in the probability of a one-time partnership (0.25 indicates a 75% reduction in one-time partnership, i.e., $\pi_{0,k} \times 0.25$), isolation scenario, vaccination start time, and vaccination scenario. The scenarios are as follows:

• Behavior change:

  • 0 = No behavior change
  • 1 = Universal behavior change (all individuals participate)
  • 2 = Targeted behavior change (only individuals in the two highest strata of sexual activity participate)

• Isolation scenario:

  • 1 = Full compliance with isolation
  • 2 = Partial compliance with isolation

• Vaccination scenario:

  • 0 = No vaccination
  • 1 = Universal vaccination availability (all individuals can recieve vaccination)
  • 2 = Targeted vaccination (only individuals in the two highest strata of sexual activity can be vaccinated)

Therefore, the file "mpox_30to110-2-0.5-1-0to-30-2.py" runs code for simulations with behavior change that begins between day 30 and day 110, behavior change scenario 2, 50% reduction in the probability of a one-time partner, isolation scenario 1, vaccination beginning on day 30, and vaccination scenario 2.

Core Functions and Input Data

• mpox_utils.py: all methods and helper functions to run the simulations
• mpox_vax_coverage_data.xlsx: file containing the number of vaccines available during each week of the simulation (see manuscript for further details)

Generate Results

• Simulations with a range of intervention timings (behavior change scenario 2, isolation scenario 2 vaccination scenario 2)

  • mpox_30to110-2-0.25-2-0to-30-2.py
  • mpox_30to110-2-0.25-2-5to30-2.py
  • mpox_30to110-2-0.5-2-0to-30-2.py
  • mpox_30to110-2-0.5-2-5to30-2.py
  • mpox_30to110-2-0.75-2-0to-30-2.py
  • mpox_30to110-2-0.75-2-5to30-2.py

• Simulations with a range of intervention timings (behavior change scenario 2, isolation scenario 1 vaccination scenario 2)

  • mpox_30to110-2-0.25-1-0to-30-2.py
  • mpox_30to110-2-0.25-1-5to30-2.py
  • mpox_30to110-2-0.5-1-0to-30-2.py
  • mpox_30to110-2-0.5-1-5to30-2.py
  • mpox_30to110-2-0.75-1-0to-30-2.py
  • mpox_30to110-2-0.75-1-5to30-2.py

• Simulations with a range of interventions and isolation scenarios

  • mpox_70-0-0.5-012-30-0.py
  • mpox_70-1-0.5-012-30-0.py
  • mpox_70-1-0.5-012-30-1.py
  • mpox_70-2-0.5-012-30-0.py
  • mpox_70-2-0.5-012-30-2.py

Data Processing and Visualization

• concatenate_simulations.ipynb: concatenates output from embarassingly parallelized computing cluster output into one file
• create_figures_main: creates plots comparing interventions and intervention timings
• create_figures_relationship_type.ipynb: creates plots showing simulation results by relationship tpe
• create_cumulative_edge_graph.ipynb: creates plots showing how edges accumulate over time in the graph

Supplement Code

This is code written to address reviewer responses and primarily creates information contained in the appendix of the manuscript.

• mpox_70-2-0.5-012-30-2_SUPPLEMENT.py - this code runs the sensitivity analyses looking at different values of the infection parameters
• mpox_N5.10.20k-70-012-0.5-1-30-012_SUPPLEMENT.py - this code runs the population size sensitivity analyses for N = 5,000, N = 10,000, and N = 20,000
• mpox_N40k-70-012-0.5-1-30-012_SUPPLEMENT.py - this code runs the population size sensitivity analyses for N = 40,000
• mpox_N80k-70-012-0.5-1-30-012_SUPPLEMENT.py - this code runs the population size sensitivity analyses for N = 80,000
• analyze_compress_supplement.py - this code helps analyze the results from the population size code and saves datasets used for the figures
• clustering_simulations.py - this code creates the information for the network structure part of the appendix
• supplement_figures.py - This code creates the new figures
• mpox_utils_supplement.py - this is a copy of the original utils file that tracks additional information during simulations required for the response to reviewers

Parameters for Simulation Scenarios

The following table shows the parameter values used in simulation and is Table 1 in the manuscript. References numbers relate to the references in the manuscript.

Network Model

Parameter	Value	Description
rt_k		Relationship type, percent of nodes (21,23)
	47.1	0 main, 0 casual
	16.7	0 main, 1 casual
	7.4	0 main, 2 casual
	22.0	1 main, 0 casual
	4.7	1 main, 1 casual
	2.1	1 main, 2 casual
π_{o,k}		Daily probability of one-time partnership formation (21,23)
	0	Stratum 1
	0.001	Stratum 2
	0.0054	Stratum 3
	0.0101	Stratum 4
	0.0315	Stratum 5
	0.286	Stratum 6
n_{o}	Geometric(1 - π_{o,k})	Number of one-time contacts per day (21)
rd_{m,e}	Geometric(1/407)	Duration of main partnership, days (21,23)
rd_{c,e}	Geometric(1/166)	Duration of casual partnership, days (21,23)

Epidemic Model

Parameter	Value	Description
β	0.9	Probability of transmission per sexual contact
π_m	0.22	Daily probability of sexual contact (main partnership) (21,23)
π_c	0.14	Daily probability of sexual contact (casual partnership) (21,23)
t_{e,k}	Normal(7, 1)	Time spent in exposed state, days (1,2,13,25-27)
t_{i,k}	Normal(27, 3)	Time spent in infectious state, days (1,2,13,25-27)

Interventions

Parameter	Value	Description
π_{b,k}	0.5	Decrease in probability of one-time partnership formation
π_{v_1}	0.14	Probability of vaccination (one dose) (32)
π_{v_2}	0.227	Probability of vaccination (two doses) (32)
VE_1	0.358	Vaccine efficacy (one dose) (34)
VE_2	0.66	Vaccine efficacy (two doses) (34)

Node-specific attributes use the subscript X_{x,k}, edge-specific attributes use X_{x,e}.