The Symbiosis Project 

Simple Epidemiological Models

 

The traditional S-I-R epidemiological models can be used as contexts for a number of mathematical, biological, and statistical investigations.  These netlogo simulations reflect that idea by exploring as many as seven different types of epidemics in a single simulation.  There are opportunities for statistical analysis and inference, for exploring, developing and implementing mathematical models, and  for addressing important biological questions.

SIR models

All of the epidemic simulations are set in a village somewhere in the world sometime near the end of the 1800's. Epidemics break out, and scientists like Charles Louis Alphonse Laveran, Ronald Ross, Carlos Finlay, and Walter Reed are trying to determine their cause.  Those real world scientists were faced with the fact that a disease can be spread in at least 7 different ways: 

  1. By direct contact with infected humans.
  2. By drinking infected water.
  3. By eating tainted food.
  4. By being bitten by flying insects like mosquitos.
  5. By breathing airborne infectious agents.
  6. By being bitten by crawling insects like ticks.
  7. By living in unsanitary housing.

Each of the simulations below implements one or more of these types of simulations.  Moreover, the human inhabitants of the village sleep for 8 hours in their houses, after which they make their way to the river for water, eat 3 times daily (food not shown), and wander about doing their daily activities. All the while, black mosquitos are flying around, including in and out of the hospital and houses. There are also orange crawling bugs, which due to their limited range do not move at all in this simulation.

Each simulation begins with 10 infecteds and 490 susceptibles (a susceptible is a person who does not have the disease but is susceptible to getting it).  When something infects an susceptible, but they don't become symptomatic for 24 hours. When they do become symptomatic, they head for the hospital. The hospital houses the symptomatics until they recover 96 hours after becoming symptomatic.

 There are variations on these themes.  For example, the Traditional SIR epidemic model is included for comparison and adheres to the strict assumptions -- uniform mixing of the population, no latency period, no hospital, etcetera. In contrast, YouBeWalterReed implements the entirety of the 7 different epidemic types listed above and is included as a challenge meant to illustrate the difficulty in identifying the specific type of an epidemic without doing extensive work beyond observation and modeling.

Links

URL Module Description
TraditonalEpidemic 1 Traditional SIR model simulation
HumanEpidemic 1 Simulation of a disease spread through Human-Human contact
MosquitoEpidemic 1 Simulation of a disease spread through Mosquito-Human contact
EpidemicVector 1 Random selection from 4 epidemic types with interventions
RandomizationTest 1 Randomization test for analyzing data collected in simulations
     
YouBeWalterReed 1 Random selection from 7 epidemic types
McastEpidemics 1 M-cast Slides from Project Mosaic M-cast
Mcast Playback 1 Playback of the Mcast
     

Other

Resources / Additional Information

  Netlogo - R interface can be used to analyze Epidemic data with R directly from a simulation. ( http://netlogo-r-ext.berlios.de/ )
  Right click and save the following Netlogo - R interface equipped version of
       TraditionalEpidemic
       HumanEpidemic
       MosquitoEpidemic
       EpidemicVector