Upload
others
View
0
Download
0
Embed Size (px)
Citation preview
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Lyme Disease: A Mathematical Approach
Antoine Marc & Carlos Munoz
July 21, 2015
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Overview
1 Biology of Lyme DiseaseBorrelia burgdoferiIxodes scapularisHosts
2 Base ModelCompartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
3 Age-Structured Tick ClassModel
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
4 Age-Structured Tick Class &Seasonality Model
System of ODEsAnalysis of R0SimulationsConclusion
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Borrelia burgdoferiIxodes scapularisHosts
Why is Lyme Disease Important to Study?
The shortening of Winter in the North led to the following:
Warmer temperatures have been predicted to both enhancetransmission intensity and extend the distribution of diseasessuch a malaria and dengue as well.
Climate change may open up previously uninhabitableterritory for arthropod vectors as well as increase reproductiveand biting rates, and shorten the pathogen incubation period.
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Borrelia burgdoferiIxodes scapularisHosts
Borrelia Burgdoferi
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Borrelia burgdoferiIxodes scapularisHosts
Ixodes scapularis
Ixodes scapularis, the black-legged tick
Can be found throughout the country including Texas
Take a blood meal every time they molt
Once infected, they are infected for life
Must attach for 36 hours to transmit the bacteria
No vertical transmission
Questing season is changing due to climate change
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Borrelia burgdoferiIxodes scapularisHosts
Figure of Life cycle:
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Borrelia burgdoferiIxodes scapularisHosts
White footed Mouse, Peromyscus leucopus
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Borrelia burgdoferiIxodes scapularisHosts
Unidentified Alternate Host
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Borrelia burgdoferiIxodes scapularisHosts
1 Recent data shows that ticks quest at two distinct heights
2 This will help narrow down the search for the other hosts
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Compartmental Model for Base Model
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Base Model
diMdt
= αβ̃M(1 − iM)iT − µM iMdiTdt
= β̃T (1 − iT )((αiM) + (1 − α)iA
)− µT iT
diAdt
= (1 − α)β̃A(1 − iA)iT − δ1µAiA
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Base Model with Seasonality
diMdt
= αβ̃M(1 − iM)iT − µM iMdiTdt
= β̃T (1 − iT )((αiM) + (1 − α)iA
)− µT iT
diAdt
= δ1(1 − α)β̃A(1 − iA)iT − δ1µAiA
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
middle2
3:61-304
δ1 =
{1 if 61 ≤ t ≤ 3040 Otherwise
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Parameters
Parameter Definition
ρM Birth rate of the mice into the susceptible classρT Birth rate of the ticks into the susceptible classρA Birth rate of the alternate host into the susceptible classβM Contact Transmission Rate for the miceβT Contact Transmission Rate for the TickβA Contact Transmission Rate for the alternate hostα Proportion of the the ticks that have a preference for questing at lower heightsµM Death rate for the Mouse classµT Death rate for the Tick classµA Death rate for the Alternate Host class
Table: Table of Variables & Parameters for Base Model
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Basic Reproduction Number of a Infection
The nondimensionalized system was reduced and rearranged intoan equation for R0, which determines whether or not there will bean epidemic.
If R0 < 1 then the disease will eventually die out of thepopulation.
If R0 = 1 the disease remains at a constant level in thepopulation.
If R0 > 1 the level of disease in the population will increaseuntil there is an epidemic.
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
R0 :=
√α2βMβT µA + (1 − α)2βAβT µM
µMµT µA
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Scenarios in which the overall R0 > 1 and an epidemic willoccur in the community.
Time (Days)
0 500 1000 1500 2000 2500 3000 3500
Pro
po
rtio
n o
f T
ota
l P
op
ula
tio
n
0
0.1
0.2
0.3
0.4
0.5
0.6
Base Model for 10 years
Infected Mouse
Infected Tick
Infected Alternate
Time (Days)
0 500 1000 1500 2000 2500 3000 3500
Pro
po
rtio
n o
f T
ota
l P
op
ula
tio
n
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
Base Model for 10 years
Infected Mouse
Infected Tick
Infected Alternate
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Overall R0 > 1 and an epidemic occurs
Time (Days)
0 500 1000 1500 2000 2500 3000 3500
Pro
po
rtio
n o
f T
ota
l P
op
ula
tio
n
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
Base Model for 10 years
Infected Mouse
Infected Tick
Infected Alternate
Time (Days)
0 500 1000 1500 2000 2500 3000 3500
Pro
po
rtio
n o
f T
ota
l P
op
ula
tio
n
0
0.05
0.1
0.15
0.2
0.25
0.3
Base Model for 10 years
Infected Mouse
Infected Tick
Infected Alternate
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Scenarios in which the overall R0 > 1 and an epidemic willoccur in the community Cont.
Time (Days)
0 500 1000 1500 2000 2500 3000 3500
Pro
po
rtio
n o
f T
ota
l P
op
ula
tio
n
0
0.05
0.1
0.15
Base Model for 10 years
Infected Mouse
Infected Tick
Infected Alternate
Time (Days)
0 500 1000 1500 2000 2500 3000 3500
Pro
po
rtio
n o
f T
ota
l P
op
ula
tio
n
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
Base Model for 10 years
Infected Mouse
Infected Tick
Infected Alternate
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Overall R0 < 1 and the disease dies out
Time (Days)
0 500 1000 1500 2000 2500 3000 3500
Pro
po
rtio
n o
f T
ota
l P
op
ula
tio
n
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1Base Model for 10 years
Infected Mouse
Infected Tick
Infected Alternate
Time (Days)
0 500 1000 1500 2000 2500 3000 3500
Pro
po
rtio
n o
f T
ota
l P
op
ula
tio
n
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
Base Model for 10 years
Infected Mouse
Infected Tick
Infected Alternate
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Conclusion
Adding an invading species with R0 > 1 can increase the levelof infection for the initial host
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Three Tick Stages
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
System of ODEs Modeling the Spread of Lyme DiseaseDiagram that incorporates Criss-Cross Infection
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
System of ODEs Modeling the Spread of Lyme Disease
diMdt
= αβM(1 − iM)iTN − µM iM (1a)diTLdt
= βTL(1 − iTL)iM − ηTL iTL − µTL iTL (1b)
diTN Edt
= βTN (1 − iTN E − iTN )(
αiM + (1 − α)iA)− ηTN iTN E − µTN iTN E
(1c)
diTNdt
= ηTL iTL − ηTN iTN − µTN iTN (1d)diTAdt
= ηTN (iTN + iTN E )− µTA iTA (1e)
diAdt
= βA(1 − iA)(iTA + (1 − α)iTN
)− µAiA (1f)
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Parameters
Parameter Definition
ρM Birth rate of the mice into the susceptible classρT Birth rate of the ticks into the susceptible classρA Birth rate of the alternate host into the susceptible Larvae classβM Contact Transmission Rate for the miceβTL Contact Transmission Rate for the larval tickβTN Contact Transmission Rate for the nymphal tickβA Contact Transmission Rate for the alternate hostηTL Rate that the larvae molt into nymphsηTN Rate that the larvae nymphs into adultsα Proportion of the the ticks that have a preference for questing at lower heightsµM Death rate for the Mouse classµT Death rate for the Tick classµA Death rate for the Alternate Host class
Table: Table of Variables & Parameters for Model with 3 Tick Classes
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Basic Reproduction Number of a Infection
R0 :=
max
{√αβMβLηTL
µM(ηTL + µTL)(ηTN + µTN ),
√βTN (1 − α)βAηTN(ηTN + µTN )µTAµA
}
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
To illustrate that the disease will be endemic in the alternate host.We chose variables in the following manner in the followingmanner:
higher βM
lower βTL , βTN , and βA
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Model with 3 Tick Classes α = 1, βM = .21, βTL =.00041, βTN = .00041, βA = .00041, R0 = 6.2214
Time (Days)
0 500 1000 1500 2000 2500 3000 3500
Pro
port
ion o
f T
ota
l P
opula
tion
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9Model with 3 Tick Classes for 10 years
Infected Mouse
Infected larvae
Exposed Nymph
Infected Nypmh
Infected Adult
Infected Alternate
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Model with 3 Tick Classes α = 1, βM = .01, βTL =.0041, βTN = .0041, βA = .0041, R0 = 2.9626
Time (Days)
0 500 1000 1500 2000 2500 3000 3500
Pro
port
ion o
f T
ota
l P
opula
tion
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
Model with 3 Tick Classes for 10 years
Infected Mouse
Infected larvae
Exposed Nymph
Infected Nypmh
Infected Adult
Infected Alternate
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
Compartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Conclusion
It’s very beneficial for the mice to be able to sustain thedisease, but not necessary.
If the the disease is endemic to the mice population, then it’svery likely that the disease will be endemic for the alternatehost population.
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
System of ODEsAnalysis of R0SimulationsConclusion
System of ODEs Modeling the Spread of Lyme Disease
diMdt
= δ3αβM(1 − iM)iTN − µM iM (2a)diTLdt
= δ1βTL(1 − iTL)iM − ηTL iTL − µTL iTL (2b)
diTN Edt
= δ3βTN (1 − iTN E − iTN )(
αiM + (1 − α)iA)− ηTN iTN E − µTN iTN E
(2c)
diTNdt
= ηTL iTL − ηTN iTN − µTN iTN (2d)diTAdt
= ηTN (iTN + iTN E )− µTA iTA (2e)
diAdt
= βA(1 − iA)(
δ5iTA + δ3(1 − α)iTN)− µAiA (2f)
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
System of ODEsAnalysis of R0SimulationsConclusion
δ1 =
{1 if active>182 active119 and active274 active41 active
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
System of ODEsAnalysis of R0SimulationsConclusion
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
System of ODEsAnalysis of R0SimulationsConclusion
Basic Reproduction Number of a Infection
R0 :=
0 0 ≤ t ≤ 121√αβMβTLηTL
(ηTL + µTL)(ηTN + ηTN )µM121 ≤ t ≤ 274
max
{√αβMβLηTL
µM(ηTL + µTL)(ηTN + µTN ),
√βTN (1 − α)βAηTN(ηTN + µTN )µTAµA
}274 ≤ t ≤ 283
0 283 ≤ t ≤ 3460 346 ≤ t ≤ 365
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
System of ODEsAnalysis of R0SimulationsConclusion
Model with Seasonality: α = 1, βM = .0011, βTL =.0011, βTN = .0011, βA = .0011, R0 = .8693
Time (Days)
0 500 1000 1500 2000 2500 3000 3500
Pro
port
ion o
f T
ota
l P
opula
tion
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1Model with 3 Tick Classes and Seasonality for 10 years
Infected Mouse
Infected larvae
Exposed Nymph
Infected Nypmh
Infected Adult
Infected Alternate
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
System of ODEsAnalysis of R0SimulationsConclusion
Model with Seasonality: α = .5, βM = .011, βTL =.011, βTN = .011, βA = .011, R0 = .7561
Time (Days)
0 500 1000 1500 2000 2500 3000 3500
Pro
port
ion o
f T
ota
l P
opula
tion
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1Model with 3 Tick Classes and Seasonality for 10 years
Infected Mouse
Infected larvae
Exposed Nymph
Infected Nypmh
Infected Adult
Infected Alternate
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
System of ODEsAnalysis of R0SimulationsConclusion
Model with Seasonality: α = 0, βM = .0011, βTL =.0011, βTN = .0011, βA = .0011, R0 = .7036
Time (Days)
0 500 1000 1500 2000 2500 3000 3500
Pro
port
ion o
f T
ota
l P
opula
tion
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1Model with 3 Tick Classes and Seasonality for 10 years
Infected Mouse
Infected larvae
Exposed Nymph
Infected Nypmh
Infected Adult
Infected Alternate
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
System of ODEsAnalysis of R0SimulationsConclusion
Conclusion
Lyme disease is found in mice in Texas at low levels. Themodel indicates that it’s very likely that there is a larger hostthat is also an effective carrier.
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
System of ODEsAnalysis of R0SimulationsConclusion
References
Branda JA, Rosenberg ES. 2013 Borrelia miyamotoi: ALesson in Disease Discovery. Ann Intern Med. 159:61-62.
Fukunaga M. 1995. Genetic and Phenotypic Analysis ofBorrelia miyamotoi sp. nov., Isolated from the Ixodid TickIxodes persulcatus, the Vector for Lyme Disease in Japan. Int.J. Syst. Bacteriol. 45: 804-810.
Rollend, L. 2013. Transovarial transmission of Borreliaspirochetes by Ixodes scapularis: a summary of the literatureand recent observations. Ticks Tick-borne Dis. 4(12):46-51.
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBase Model
Age-Structured Tick Class ModelAge-Structured Tick Class & Seasonality Model
System of ODEsAnalysis of R0SimulationsConclusion
References Cont.
Anderson, J., L. Magnarelli. 1980. Vertebrate hostrelationships and distribution of ixodid ticks (Acari: Ixodidae)in Connecticut, USA. Journal of Medical Entomology,17:314-323.
Bertrand, M., M. Wilson. 1996. Microclimate-dependentsurvival of unfed adult Ixodes scapularis (Acari: Ixodidae) innature: life cycle and study design implications. Journal ofMedical Entomology. 33: 619-627.
Antoine Marc & Carlos Munoz Lyme Disease: A Mathematical Approach
Biology of Lyme DiseaseBorrelia burgdoferiIxodes scapularisHosts
Base ModelCompartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Age-Structured Tick Class ModelCompartmental ModelSystem of ODEsAnalysis of R0SimulationsConclusion
Age-Structured Tick Class & Seasonality ModelSystem of ODEsAnalysis of R0SimulationsConclusion