I Math 1LS3 Assignment 52

Math functions in context; Chapters 2-5 (geese) 1-4 (elephants)

1. The following excerpt is taken from Connection between oligomeric state and gating
characteristics of mechanosensitive ion channels. Christoph A. Haselwandter and Rob
Phillips. PLoS Computational Biology. 9.5 (May 2013)
- - -
The central quantity in this model is the channel opening
probability

1
Pa= l +eP(M'-,AA} • (l)

where P= 1/kBT, in which kB is Boltzmann's constant and Tis

the temperature, ag is the total free energy difference between the
open and closed states of MscL, r is the membrane tension, and
AA is the area difference between the open and closed channel
states.

From the context we figured out that the membrane tension T and the parameter j3 are
positive numbers.
(a) Consider Po as a function of the membrane tension; what is its domain?

(b) Probability is a number between 0 and 1 (representing the range from 0% to 100%).
Show that Po is indeed a probability.

(c) Assume that .6.A > 0. What is the limit of Po as the membrane tension keeps increasing
(i.e., as the membrane tension approaches infinity)?

Continued on next page

267
 2. The following excerpt is taken from Evaluating the adequacy of gravity models as
a description of human mobility for epidemic modelling. James Truscott and Neil M.
Ferguson. PLoS Computational Biology. 8.10 (Oct. 2012)
By setting parameters to zero,we can examine a ·range· of sub-
mpdek. We look at models of distance interaction oftwo fonns,
smooth· kemel{SK)

1[ d=O,
f(d).= {
1
d>O.
(l +d/r:1.)1

The exponent in the denominator in the second line is 1 .

(a) Assume that a> 0 and 1 > 0. What is the limit of the kernel f(d) as d---+ oo?

(b) Assume that a> 0 and 1 < 0. What is the limit of f(d) as d---+ oo?

(c) What is the limit of the kernel f (d) as d ---+ o+?

(d) Assume that f(d) = 7r ford~ 0. Is the kernel f(d) continuous at d = O?

Continued on next page

268
Math 1LS3 Assignment 52

3. The following excerpt is taken from The laminar cortex model: a new continuum
cortex model incorporating laminar architecture. J. Du, V. Vegh, and D.C. Reutens. PLoS
Computational Biology. 8.10 (Oct. 2012).
the average of membrane potentials of neurons in the
element, that is

V= NeVe+NiVi
Ne+Ni
where Ne}Ni are the numbers of excitatory and inhibitory neurons
and V.., and Vi are the (average) membrane pot-entlals of exdtatory
and inhibitory neuron populations respectlvely.
(a) View V as a function of ¼ (in which case all remaining symbols on the right side
represent parameters). Describe in words the graph of V.

(b) View V as a function of Ni (in which case all remaining symbols on the right side
represent parameters). What is the limit of V as Ni increases beyond any bounds (i.e.,
approaches oo)?

(c) View Vas a function of Ve. What is the limit of Vas½ approaches -oo?

Continued on next page

269
4. Some models involving Allee effect incorporate the fact that bacteria produce poorly
when the population size is small. One such model states that

2 7a ba+l
b - . t
t+l - 1.3 + b;
where the parameters a and a is positive. The population bt is measured in millions. So
bt = 2 means 2 million bacteria.
(a) Identify the per capita production function.

(b) Write down the equation you would need to solve to find the equilibria of the given
system. Simplify as much as possible.

(c) Solve the equation in (b) for the values a = 1 and a = 2. Make sure to eliminate those
solutions which do not make sense.

Continued on next page

270
Math 1LS3 Assignment 52

5. In studying L-form bacteria subpopulation in human gut researchers use the model

bt+l = bt ( ae-mbt - de-nbt + l)

where a, d, m, n > 0 and n - m > 0.

(a) Identify the per capita production function.

(b) For which values of the parameters a and d does the equilibrium make (biological)
sense?

Continued on next page

271
 6. The following excerpt is taken from Predictive dynamics of human pain perception.
Guillermo A. Cecchi et al. PLoS Computational Biology. 8.10 (Oct. 2012).

Ove:rfitting of· the model was investigated ·using the .Akaike

Information Criterion (AlC), which penalizes the measure ·of
goodness.· of fit with a term proportional to the number of ·tree
parameters [!HJ. When the .residual squared error sum (&.\) is
known, the criterion •can be written as

AIC-,Hog (nss) +2k+C

where ,a is the number of samples, and k the number of
parameters. C is a constanl

Recall the convention log= log 10 . Assume that SS > 0.

(a) Find the rate of change of AIC with respect ton.

(b) Find the limit of AIC as the number of samples n approaches oo.

Continued on next page

272
Math 1LS3 Assignment 52

7. The generalized linear-quadratic (LQ) model for the percent S of cancer cells surviving
radiation treatment states that

S(d) = exp (-and - (3nd 2 + ,T)

where d > 0 is the dose (in Gray) per treatment of radiation, n is the number of treatments
applied (also called fractions), 1 is the tumour cell repopulation rate (since new tumour
cells grow during treatment) and T > 0 is the total treatment duration. The parameters
a, (3 and I are positive. Recall that exp(x) = ex.
(a) Assume that 1 = 0. What is the range of Sin that case?

(b) Explain why S(d) is a continuous function ford> 0.

(c) Let f (d) = -and - (3nd 2 + 1 T (i.e., f is the function in the exponent). Assume that
1 = 0. Make a sketch of f(d).

Continued on next page

273
(d) Make a sketch of f(d) in a general case when when 1 > 0.

(e) Find all critical points of S (d).

THE END

274
I Math 1LS3 Assignment 53

Math functions in context; Chapters 5,6 (geese) 4,5 (elephants)

1. The following excerpt is taken from Modeling of pharmacokinetics of cocaine in human

reveals the feasibility for development of enzyme therapies for drugs of abuse. Chang-Guo
Zhan and Fang Zheng. PLoS Computational Biology. 8.7 (July 2012).

A Pharmacoldnedc model - becm ·developc:d to examine the

effects of ;i cocaine-metabolizing enzyme in plasma on the ·time
0.003-------------

!0.002
&
Jo.~
.c
8 0

0 20 40 60 80
Tlme(mln)
(a) Consider the function f(t) = Ate-f3t whose graph is shown in Figure 3.3.41d in your
textbook. Find the relative maximum of f(t) and the value oft where it occurs.

Continued on next page

275
(b) In the given concentration vs time graph estimate the maximum concentration and
the time when it occurs. Use these estimates to find A and /3, thus obtaining a formula
(model) for the concentration as a function of time.

(c) By imitating (a), and (b), obtain another model (i.e., find A and k) for the concentra-
At
tion, based on the function c(t) = k with A, k > 0 (Figure 3.3.41.e).
+t2

Continued on next page

276
Math 1LS3 Assignment 53

2. The following excerpt is taken from Hybrid equation/agent-based model of ischemia-

induced hyperemia and pressure ulcer formation predicts greater propensity to ulcerate in
subjects with spinal cord injury. Alexey Solovyev et al. PLoS Computational Biology. 9.5
(May 2013).
Based on this expression of J-1'.~, an
explicit solution for / 2(~ can be derived with initial conditions
11(0) = J2(O) = 0. This solution has the following form

Here J.1.;resu a, b, pi, and P2 are constants expressed in terms of Ri,

R2, R3, Rt, C1, ½, Vo,

To avoid exponents (which are written in smaller font, and thus could be hard to read),
we sometimes use exp( x) for the exponential function ex.
Assume that 12 rest = 1, Pl = 2 and P2 = 2.4. Find all critical points of h(t). (Your answer
will contain a ~nd b).

Continued on next page

277
 3. The following excerpt is taken from Host resistance, population structure and the long-
term persistence of bubonic plague: contributions of a modelling approach in the Malagasy
focus. Fanny Gascuel et al. PLoS Computational Biology. 9.5 (May 2013).

Using the·Next Generation Method,

we found

(3)

The propagation of plague is favoured by .a high trartsrrussfon

from fl~a:i to rats (13); which increase& the number of infectious rats,
and by a hlgh flea carrying capacity of rats (K1), which increases
the number of free fleas, the vectors ofthe disease (equation (3}).. It
is also favoured by a hlgh tarrying capacity of the rat population,
K, aud by a .high. search efficiency of fleas, a, thrQugh their direct
effect on the.probability 1- e-aK that a flea finds a host.
By saying "It (meaning the propagation of plague Ro) is also favoured by a high carrying
capacity of the rat population, K, ... " the authors meant to say that the propagation of
plague Ro is an increasing function of the rat population K. Prove that this statement is
correct. (All parameters are assumed to be positive.)

Continued on next page

278
Math 1LS3 Assignment 53

4. The generalized linear-quadratic (LQ) model for the percent S of cancer cells surviving
radiation treatment states that

S = exp (-and - f3nd 2 + 1T)

where d > 0 is the dose (in Gray) per treatment of radiation, n is the number of treatments
applied (also called fractions), 'Y is the tumour cell repopulation rate (i.e., new tumour cells
growing during treatment) and T > 0 is the total treatment duration. The parameters a,
/3 and 'Y are positive. Recall that exp( x) = ex.
(a) Assume that 'Y = 0 and view S as a function of the dose d. Show that S is a decreasing
function. Does it make sense?

(b) Assume that /3 = 0 and view S as a function of the dose d. Show that S a decreasing
function.

Continued on next page

279
(c) Assume that a = 0 and view S as a function of the dose d. Show that S a decreasing
function.

(d) View S as a function of the number of treatments applied n. Show that Sis a decreasing
function. Interpret your answer.

Continued on next page

280
Math 1LS3 Assignment 53

5. The interstitial fluid pressure (IFP) p the at a location r mm from the centre of a
tumour is given by
1.3 sinh(0.4r)
p = 0.26 7Pi+---.- -
r smha
where Pi is the atmospheric pressure (assumed constant) and a is a positive parameter
which involves the values related to a fluid movement within the tumour.
Searching Wikipedia, you found that sinh x = ½(ex - e-x) .
The researchers claim that p is an increasing function of r when r > 2.5. Justify their
claim.

Continued on next page

281
 6. The following excerpt is taken from Evaluating the adequacy of gravity models as
a description of human mobility for epidemic modelling. James Truscott and Neil M.
Ferguson. PLoS Computational Biology. 8.10 (Oct. 2012)
By setting parameters to zero,we can examine a ·range· of sub-
mpdek. We look at models of distance interaction oftwo fonns,
smooth· kemel{SK)

1[ d=O,
f(d).= {
1
d>O.
(l +d/r:1.)1

The exponent in the denominator in the second line is 1 .

(a) Find and simplify the formula for the rate of change of the smooth kernel f(d) with
respect to din the case when d > 0.

(b) Find and simplify f" (d).

THE END

282
I Math 1LS3 Assignment 54

Math functions in context - Differentiation

1. The following excerpt is taken from Connection between oligomeric state and gating
characteristics of mechanosensitive ion channels. Christoph A. Haselwandter and Rob
Phillips. PLoS Computational Biology. 9.5 (May 2013)
- - -
The central quantity in this model is the channel opening
probability

I
Po= l +eP(Ag-'[AA} • (l)

where /J= 1/kBT, in which kB is Boltzmann's constant and Tis

the temperature, ag is the total free energy difference between the
open and closed states of MscL, f is the membrane tension, and
AA is the area difference between the open and closed channel
states.

From the context we figured out that all parameters are positive numbers.
Show that Po is an increasing function of T.

Continued on next page

283
 2. The following excerpt is taken from The laminar cortex model: a new continuum
cortex model incorporating laminar architecture. J. Du, V. Vegh, and D.C. Reutens. PLoS
Computational Biology. 8.10 (Oct. 2012).
the average of membrane potentials of neurons in the
element, that is

V= NeVe+NiJ'i.
Ne+Ni
where Nei N 1 are the numbers of excitatory and inhibitory neurons
and Ve and Vi are the (average) membrane potentials of excitatory
and •inhibitory neuron populalions respectlvely.

You know that the numbers Ne and Ni are positive, and the membrane potentials½ and
¼ are negative.
(a) Assume that Vis a function of½. Find its derivative and interpret your answer.

Continued on next page

284
Math 1LS3 Assignment 54

For (b) and (c) assume that Vis a function of Ne.

(b) Find the derivative of V and interpret your answer.

(c) Find the relation between ½ and ¼ which makes the graph of V concave up.

Continued on next page

285
3. In the paper Evaluating the adequacy of gravity models as a description of human mo-
bility for epidemic modelling. James Truscott and Neil M. Ferguson. PLoS Computational
Biology. 8.10 (Oct. 2012), we read the following:

Comparison of the behaviour of the epidemic model on

different networks iA based ort the times to first infection for
network. nodes from a given initial :infectiQn site. Under certain
simplifying ass:umptions, an approximation for the mean time to
infection between two nodes can be calculated for the above
epidemfological model.

The parameters A1,2 and i; 1,2 represent network properties; such as

the fraction ofjourneys between the two nodes. r= fJxFPw+rri& tl'ie
epidemic growth rate in alatge node. The equation illustrates how

Assume that A2 = 0, and that the approximately equals sign is actually an equals sign.
Find the rate of change of t with respect to 1 .

Continued on next page

286