Biology



--------------------------------------------------------------------------------


Content:

Bubonic Plague model: which strategy is better immunization or improved medicines ? 
Monte Carlo simulation of the Soho Cholera epidemic of 1848. 
The genetics of populations: the Hardy-Weinberg ratio 


--------------------------------------------------------------------------------


A mathematical model of an epidemic of Bubonic Plague

Objective: the student should be able to decide which medical treatment is better.

Introduction

Some people consider mathematics the field of a small group of specialist and 
they think they don't need it. In this article I will give a description of 
medical doctors in a developing country. As you may know, epidemics are likely 
to occur when large groups of people gather together under bad circumstances, 
as they are likely to occur in a civil war. Medics send to refugee camps know 
about their diseases and medicine, but do they know enough of mathematics? Try 
to picture yourself in such a situation and try to answer the question: "how can 
I help these refugees most effectively, when an epidemic occurs". There are two 
possible options: 1: immunize the population or 2: use better medical treatments 
to help your patients recover from the disease. Ofcourse there is always a 
problem: lack of sufficient money! This means you'll have to make choices, you 
can spend the money only once. In the following article we will first arrive at 
a proper model to describe the population and the epidemic. Then we will 
simulate the epidemic with the help of a graphic calculator program (ApLet) to 
decide which option is best in several situations. You will realise that medical 
doctors in the field don't have such a calculator; they will have to make 
educated guesses.

The Model

The population can be subdivided in four different groups at time t, measured in 
days:

x1(t) = people who are susceptible to the disease
x2(t) = people who are currently infected with it
x3(t) = people who have been immunized against it
x4(t) = people who have already died from it
In order to analyse the changes in the above four groups, it is assumed that 
each day a fraction

a of susceptibles are immunized
b of susceptibles are infected
c of infectives die
d of infectives are cured and immunized

Let us assume that the new susceptibles and infectives enter the population at 
input rates U1(N+1) = [ 1 - (a+b)]U1(N)
This equation states that the new number of susceptibles is equal to the number 
of susceptibles the day before minus the fraction 'a' times this number minus 
the fraction 'b' times this number, meaning the number of people that have 
become immunized or infected respectively.

Now we can state the equations for all separate groups. We can do this in either 
of three ways: using differential equations, using matrices or using difference 
equations. Since we are interested in discrete periods of time and a graphical 
representation we will use difference equations. We can utilize the sequence 
Aplet to model the population. The equations are already in the ApLet, as are 
the different parameters. Eisen proposes to set the following values to the 
parameters and variables:


x1(0) = 10000 
x2(0) = 100 
x3(0) = 0 
x4(0) = 50 
a = 0.10 
b = 0.05 
c = 0.08 
d = 0.08 
Medical help makes it possible to increase a to 0.15 or d to 0.10 (b and c 
cannot be altered). Assume that the cost of increasing a parameter is linear. 
This means that increasing a one unit is twice as expensive as increasing d one 
unit. It is not possible to decrease either parameter under the specified level.
Simulate with this model and derive a conclusion which action is best. Another 
way of deriving a conclusion is looking at the model (draw a graph) and use your 
intuition.

Does it matter when the buget relation is not linear anymore? 

Bubonic Plague ApLet (Dutch Version with WordPerfect documents) 
