Multi-strain persistence induced by host age structure
|
|
- Jeremy O’Connor’
- 6 years ago
- Views:
Transcription
1 Multi-strain persistence induced by host age structure Zhipeng Qiu 1 Xuezhi Li Maia Martcheva 3 1 Department of Applied Mathematics, Nanjing University of Science and Technology, Nanjing, 194, P R China Department of Mathematics, Xinyang Normal University, Xinyang, 464, P R China 3 Department of Mathematics, University of Florida, Gainesville, FL ,USA Abstract. In this paper, we formulate an age-structured epidemic model with two competing strains. The model incorporates disease-induced mortality so that the population can not be assumed to be in a stationary demographic state. We derive explicit expressions of the basic and invasion reproduction numbers for strain one and two, respectively. Analytical results of the model show that the existence and local stability of boundary equilibria can be determined by the reproduction numbers to some extent. Subsequently, under the condition that both invasion reproduction numbers are larger than one, the coexistence of two competing strains is rigorously proved by the theory of uniform persistence of infinite dimensional dynamical systems. However, the results for the corresponding age-independent model show that the two competing strains can not coexist. This implies that age structure can lead to the coexistence of the strains. Numerical simulations are further conducted to confirm and extend the analytic results. Key Words. Epidemic model with two competing strain; age structure; coexistence; reproduction number; competition. Mathematics subject classifications(): 9D3, 9D5, 35L6 1 Introduction Epidemiological studies have shown that the susceptibility to many diseases often varies with age. For example, children of school age usually exhibit higher susceptibility to influenza than adults [1]. Host age structure is often an important factor which affects the dynamics of disease transmission. In this article we incorporate age structure into an epidemic model and study its impact on the competition of two strains. During the past decades, the age-structured epidemic models have been extensively studied by many authors (see, for example, [, 3, 4, 5] and references therein). These studies have enriched nustqzp@mail.njust.edu.cn xzli66@sina.com maia@ufl.edu 1
2 our knowledge of epidemic models with age-structure. However, most of the models neglect the disease-induced death rate so that the population can be assumed to be in a stationary demographic state. It is this assumption that simplifies significantly the analysis of the models. Actually, the disease-induced mortality is inevitable component of many diseases, especially deadly infectious diseases, such as pandemic influenza [6], West Nile virus[7], dengue virus[8], HIV and others. Investigating the impact of the disease-induced mortality on the dynamics of diseases, particularly multi-strain diseases is a neglected activity of significant importance. Few papers have investigated the impact of host age and disease-induced mortality on the dynamics of multi-strain interactions. In reality, however, many of the diseases which are still significant public health problem, are caused by more than one antigenically different strains of the causative agent [9]. These diseases include killer-diseases such as malaria, tuberculosis and HIV/AIDS. Therefore, it is necessary to study age-structured epidemic model with multiple strains. In this paper we formulate an age-structured model with two competing strains, and we incorporate the disease-induced mortality into the model. Then we study the dynamics of the model, and further investigate how the age-structured affects the interactions between two competing strains. The dynamics of the pathogen-host interactions involving multiple strains has fascinated researchers for a long time [9]. The competitive exclusion principle is a classic result in this field, which states that no two species can indefinitely occupy the same ecological niche [1]. Using a multiple-strain ODE epidemic model, Bremermann and Thieme [11] proved that the principle of competitive exclusion is valid under the assumption that infection with one strain precludes additional infections with the other strains. However, it is a common phenomenon that multiple strains coexist in nature. For instance, dengue fever has four different serotypes, often coexisting in the same geographical region [9]. It follows from the competitive exclusion principle that there must be some heterogeneity in the ecological niche. Identifying the factors that allow multiple strains to coexist is an important topic in theoretical biology. Studies have pointed out to a number of mechanisms, such as superinfection [8, 1], co-infection [13, 16], partial cross-immunity [17], density dependent host mortality [18], that can lead to coexistence of strains. In a recent article Martcheva et. al. [14] suggested that host age-structure coupled with disease-induced mortality may lead to coexistence of competing pathogens. This result was obtained numerically. In this article we investigate a similar scenario but we prove rigorously that if both invasion reproduction numbers are larger than one the two strains will persist in the age-structured model. To our knowledge this is the first result on persistence of multiple strains in a partial differential equation model. Persistence of a single pathogen has been addressed in multiple articles and it follows from the fact that the reproduction number is greater than one [15]. This paper is organized as follows. In the next section, we introduce our age-structured epidemic
3 model with two competing strains. In Section 3 we consider the age independent case. In Section 4, we discuss the existence of equilibria for the age-structured model and investigate the local stability of boundary equilibria. Section 5 is devoted to deriving sufficient conditions for the persistence of the two competing strains. In section 6, we present several numerical simulations which support and extend our theoretical results. In the conclusions (section 7) we discuss our results. Model formulation In this section, we formulate an age structured epidemic model with two competing strains. We start with a typical Gurtin-MacCamy model which describes the dynamics of an age-structured population without disease. Let n(a, t) denote the age density of the total population at age a and time t. Without the disease, we always assume that n(a, t) is described by the following Gurtin-MacCamy equation [19]: n(a, t) t + n(a, t) n(a, ) = n (a), = µ(a)n(a, t), (.1) In (.1), n (a) is a given function. n(, t) = RΦ(Q(t)) d β(a)n(a, t)da. Furthermore, a + represents the maximum life span of the individuals; µ(a) is the age-specific death rate, which satisfies µ( ) L 1 loc (, a+ ), µ(a) in (, a + ) and µ(σ) = + ; β(a) is the age-specific per capita birth rate, with β( ) L (, a + ), β(a) in (, a + ) and β(a) normalized to satisfy β(σ)π(σ)dσ = 1. (.) In the expression above π(σ) is the survival probability which gives the probability at birth of surviving to age σ. The probability of survival is defined as π(σ) = e σ µ(τ)dτ, σ (, a + ). Q(t) is a weighted average of the population density defined as Q(t) = r(σ)n(σ, t)dσ, where r(σ) is a weight kernel and r( ) L (, a + ), r(σ), σ (, a + ). Furthermore, Φ(x) is a function describing density-dependence of births and satisfying the following properties: (P1) Φ(x) is continuously differentiable; 3
4 (P) Φ(x) is strictly decreasing; (P3) Φ() = 1, lim x + Φ(x) =. Finally, R d is the demographic basic reproduction number. Now we are able to formulate our model. In order to formulate an age-structured epidemic model with two competing strains, we need to introduce some additional notation. We consider an age-structured SI epidemic model so that the population n(a, t) is divided into three classes: susceptible, infected with strain one, and infected with strain two. Let s(a, t), i 1 (a, t) and i (a, t) denote the associated density functions with these respective epidemiological age-structured classes, then we have n(a, t) = s(a, t) + i 1 (a, t) + i (a, t). We assume that all newborns are susceptible and the extra-mortality at age a due to the infection of strain j is γ j (a), j = 1,. We take the transmission rate λ j (a, t), j = 1, in the separable inter-cohort constitutive form for the force of infection generated by i j (a, t): λ j (a, t) = K j (a) q j (σ)i j (σ, t)dσ, j = 1,, where q j (a) is the age-specific infectiousness for strain j, K j (a) is the age-specific susceptibility of susceptible individuals, and q j (a), K j (a) in [, a + ], q j ( ), K j ( ) L (, a + ). Based on the above assumption, the joint dynamics of the age-structured epidemiological model are governed by the following partial differential equations s(a, t) + t i 1 (a, t) t i (a, t) t s(a, t) + i 1(a, t) + i (a, t) with initial and boundary conditions = (µ(a) + λ 1 (a, t) + λ (a, t))s(a, t), = λ 1 (a, t)s(a, t) (µ(a) + γ 1 (a))i 1 (a, t), = λ (a, t)s(a, t) (µ(a) + γ (a))i (a, t), (.3) s(, t) = R d Φ(Q(t)) β(a)n(a, t)da, i 1 (, t) = i (, t) = ; s(a, ) = s (a), i 1 (a, ) = i 1 (a), i (a, ) = i (a). (.4) In the following sections we mainly analyze the dynamics of system (.3) to investigate how the age-structure affects the interactions between the two competing strains. In order to study this question, we consider in the next section the dynamical properties of the age-independent model. 4
5 3 The age-independent case In this section we present some results on model (.3) for the age independent case. Here we always assume that a + =. If the parameters µ(a), β(a), r(a), K i (a), q i (a), γ i (a), j = 1, in the model (.3) do not depend on age, i.e, µ(a) µ, β(a) β, r(a) r, K i (a) K i, q i (a) q i, γ i (a) γ i, j = 1,, then condition (.) implies that β = µ, and without the disease, i.e., i 1 (a, t) = i (a, t) =, the corresponding equation for the total population size N(t), where N(t) = n(a, t)da, can be obtained by integrating (.1) from a = to a +, using the initial condition, and n(a +, t) =. The last condition simply means that no individual in the population can survive to age a +. Then the total population N(t) develops according to the equation dn(t) dt = R d Φ(rN(t))βN(t) µn(t). (3.1) It follows from the properties (P1)-(P3) that if R d 1 all solutions of the system (3.1) converge towards zero, and if R d > 1 all solutions of (3.1) converge towards 1 r Φ 1 ( 1 ). In fact, 1 R d r Φ 1 ( 1 ) R d is the environmental carrying capacity for the population. Let S(t) = s(a, t)da, I 1 (t) = i 1 (a, t)da, I (t) = i (a, t)da. Then the corresponding equations for S(t), I 1 (t) and I (t) can also be obtained by integrating (.3) from a = to a +, and using the conditions (.4) and s(a +, t) = i 1 (a +, t) = i (a +, t) =. The resulting equations are ds(t) = R dt Φ(rN(t))βN(t) d µs(t) K 1 q 1 I 1 (t)s(t) K q I (t)s(t), di 1 (t) = K 1 q 1 S(t)I 1 (t) (µ + γ 1 )I 1 (t), dt di (t) = K q S(t)I (t) (µ + γ )I (t). dt It is easy to see that system (3.) is the standard SI epidemic model with two strains. The system (3.) has been analyzed in [11]. The reproduction number of system (3.) for strain j, j = 1, is established in paper [11], which can be expressed as R j = K jq j 1 µ + γ j r Φ 1 ( 1 R d ). By the results in [11], we can obtain the following theorem. 5 (3.)
6 Theorem 3.1. The disease-free equilibrium E = ( 1 r Φ 1 ( 1 ),, ) of system (3.) is globally R d asymptotically stable if R 1 1 and R 1. If R j > 1 for at least one j {1, }, then the strain with the larger reproduction number uniformly persist and the other one dies out. Coexistence of the two strains is no possible except for the degenerate case R 1 = R. Theorem 3.1 implies that if the system (.3) does not depend on the age structure the competitive exclusion principle holds, i.e, two competing strains can not coexist. In the following sections we will prove that the coexistence of two competing strains is possible if the system depends on age structure. Accordingly, this indicates that the age-structure of the host is one of the mechanisms which can lead to the coexistence of two competing strains. 4 Boundary equilibria and local analysis We now study the dynamics of the model (.3) to investigate the impact of the age structure on strain competition and coexistence. We start with some notations and some results on the subsystem of model (.3). Let L 1 +(, a + ) = {φ L 1 (, a + ) : φ(a) R 1 + for almost all a (, a + )}; ã = min{a : a β(σ)dσ = }; β max = ess β min = ess q j min = ess K j min = ess r min = ess sup β(a); a (,a + ) inf β(a); a (,a + ) inf q j(a); a (,a + ) inf K j(a); a (,a + ) inf r(a); a (,a + ) x y := max{x, y}, x, y R; x y := min{x, y}, x, y R. The relation ϕ φ, ϕ, φ L 1 +(, a + ) indicates that ϕ(a) φ(a) for almost all a (, a + ). The biological interpretation of the variables requires us to consider the phase space of the model (.3) to be the space X + := {(s ( ), i 1 ( ), i ( )) : s ( ) L 1 +(, a + ), i j ( ) L 1 +(, a + ), j = 1, }. By standard methods [], it is possible to prove existence and uniqueness of solutions to system (.3). Moreover, it is easy to show that all the solutions remain nonnegative and bounded for t >. Furthermore, in the following two sections, we always assume that 6
7 (H1) There exists ã < ã such that β(a) > if a (ã, ã); (H) K j (a) >, q j (a) >, j = 1, for all a [, a + ) The dynamics of the Gurtin-MacCamy system has been extensively studied by many authors [19, 1, ]. In the paper, in order to ensure that the system (.1) is dissipative we assume the following: (H3) r min > and there exists a constant M > such that Φ(r min x)β max x < M for all x [, + ). Then we have the following theorem: Theorem ) If R d ) Assume R d > 1. Then i) if ã n (a)da = we have < 1, then lim t + n(, t, n (a)) = for each n ( ) L 1 + (, a+ ). lim n(, t, n (a)) = ; t + ii) if (H1),(H3) hold, then there exists a constant ɛ >, which does not depend on the initial conditions, such that every solution n(a, t) with initial condition n ( ) Γ satisfies where lim inf n(a, t, n ( ))da ɛ, t + Γ := {n ( ) L 1 + (, a+ ) : ã n (a)da > }. The proof of Theorem 4.1 can be found in the Appendix. In the case R d < 1, Theorem 4.1 tells us that the global behavior of system (.1) is clear: the population dies out. So in the following we always assume that R d > 1. When Rd > 1, it is easy to see that system (.1) has a unique nontrivial equilibrium, given by n (a) = π(a) r(σ)π(σ)dσ Φ 1 ( 1 R d ). In order to make the mathematics tractable, we make the following additional assumption: (H4) If R d > 1, then the unique nontrivial steady state n (a) of system (.1) is globally asymptotically stable in Γ. Now let us consider the dynamical properties of a reduced system of system (.3) in which only 7
8 one strain is present. The reduced system is governed by the equations: s(a, t) s(a, t) + = (µ(a) + λ j (a, t))s(a, t), t i j (a, t) + i j(a, t) = λ j (a, t)s(a, t) (µ(a) + γ j (a))i j (a, t), t s(, t) = R d Φ(Q(t)) β(a)(s(a, t) + i j (a, t))da, i j (, t) =, s(a, ) = s (a), i j (a, ) = i j (a). (4.1) The existence and local stability of endemic equilibria for the subsystem (4.1) has been analyzed in paper [3] where γ j (a) is constant, but the analysis method can be used to analyze the subsystem (4.1). As in paper [3], we define the basic reproduction number for strain j, j = 1, as Then we have R j = a+ K j (a)n (a) q j (σ)e σ a (µ(θ)+γ j (θ))dθ dσda, j = 1,. a Theorem 4.. Let R d > 1 and (H1)-(H4) hold. If R j > 1, the disease free equilibrium Ẽ := (n (a), ) of system (4.1) is unstable, and 1) if ã s (a) + i j (a)da =, then as t + we have s(, t, (s ( ), i j ( ))), i j (, t, (s ( ), i j ( ))) ; ) if ã s (a) + i j (a)da > and i j (a) for almost all a (, a + ), then as t + we have s(, t, (s ( ), i j ( ))) n ( ), i j (, t, (s ( ), i j ( ))) ; 3) otherwise, (s ( ), i j ( )) Γ j and there exists a constant δ > such that every solution (s(a, t), i j (a, t)) with initial condition (s ( ), i j ( )) Γ j satisfies where lim inf t i j (a, t)da δ, Γ j := {(s ( ), i j ( )) L 1 + (, a+ ) L 1 + (, a+ ) : i j (a)da >, ã (s (a) + i j (a))da > }. The first and second conclusions can be proved by using a similar way as in the proof of Theorem 4.1. Similar to the proof of Theorem 4.1 we can also prove the third conclusion of Theorem 8
9 4.. Here, we omit the detailed proofs of the theorem, and they are left to the reader. It follows from the results in paper [3] that the system (4.1) has at least one positive equilibrium if R j > 1. In this paper we always assume that (H5) If R j > 1, the system (4.1) has a unique positive equilibrium, denoted by Ẽ := (s j (a), i j (a)), which is globally asymptotically stable in Γ j. It is easy to see that if R d > 1 and R j > 1, j = 1, the system (.3) has four boundary equilibria which we label as E = (,, ), E = (n (a),, ), E 1 = (s 1 (a), i 1 (a), ), E = (s (a),, i (a)). In order to study the local stabilities of the boundary equilibria, let us introduce the invasion reproduction number for each of the strains. The invasion reproduction number of the strain j, j = 1, measures the ability of the strain j to invade an equilibrium of the strain k, i = 1,, k j. We define the invasion reproduction of the strain j as Then we have a+ R j k = K j (a)s k (a) a q j (σ)e σ a (µ(θ)+γ j (θ))dθ dσda, j, k = 1,, j k. Theorem 4.3. Assume R d > 1, R j > 1, j = 1, and (H5) holds. Then (1) E, E are always unstable; () E 1 is locally stable if R 1 < 1, and E 1 is unstable if R 1 > 1; (3) E is locally stable if R 1 < 1, and E is unstable if R 1 > 1. Proof. The proof of the first conclusion is relatively simple, and it is left to the reader. In the following we need to prove the second and third conclusions. Here we only pick the second conclusion to prove, and the third conclusion can be proved in a similar way. Linearizing the system (.3) about E 1, by defining the perturbation variables x 1 (a, t) = s(a, t) s 1 (a), x (a, t) = i 1 (a, t) i 1 (a), x 3(a, t) = i (a, t), 9
10 we obtain the following system x 1 (a, t) t x (a, t) t x 3 (a, t) t + x 1(a, t) + x (a, t) + x 3(a, t) = ( µ(a) + K 1 (a) = K 1 (a) = K (a) s 1(a) K j (a) j=1 q 1 (σ)i 1 (σ)dσ) x 1 (a, t)+ q j (σ)x j+1 (σ, t)dσ, q 1 (σ)x (σ, t)dσs 1 (a) + K 1(a) q 1 (σ)i 1(σ)dσx 1 (a, t) (µ(a) + γ 1 (a))x (a, t), x 1 (, t) = R d Φ(Q 1) x (, t) =, x 3 (, t) =. q (σ)x 3 (σ, t)dσs 1 (a) (µ(a) + γ (a))x 3 (a, t), β(σ)(x 1 (σ, t) + x (σ, t) + x 3 (σ, t))dσ +R d Φ (Q 1 ) β(σ)(s 1 (σ) + i 1 (σ))dσ where Q 1 = r(σ)(s 1 (σ) + i 1 (σ))dσ. Let r(σ)(x 1 (σ, t) + x (σ, t) + x 3 (σ, t))dσ, (4.) x 1 (a, t) = x 1(a)e λt, x (a, t) = x (a)e λt, x 3 (a, t) = x 3(a)e λt, (4.3) where x 1 (a), x (a), x 3 (a) are to be determined. Substituting (4.3) into (4.), we obtain 1
11 λx 1 (a) + x 1 (a) = ( µ(a) + K 1 (a) s 1(a) K j (a) j=1 q 1 (σ)i 1 (σ)dσ) x 1 (a)+ q j (σ)x j+1(σ)dσ, x 1 () = Rd Φ(Q 1 ) β(σ)(x 1 (σ) + x (σ) + x 3 (σ))dσ + Rd Φ (Q 1 ) λx (a) + x (a) x () =, λx 3(a) + x 3 (a) x 3 () =. β(σ)(s 1(σ) + i 1(σ))dσ = K 1 (a) r(σ)(x 1(σ) + x (σ) + x 3(σ))dσ, q 1 (σ)x (σ)dσs 1 (a) + K 1(a) q 1 (σ)i 1(σ)dσx 1(a) (µ(a) + γ 1 (a))x (a), = K (a) q (σ)x 3(σ)dσs 1(a) (µ(a) + γ (a))x 3(a), (4.4a) (4.4b) (4.4c) Integrating the first equation of (4.4c) from to a yields x 3(a) = a K (σ)s 1 (σ)e a a + σ (λ+µ(θ)+γ(θ))dθ dσ q (a)x 3(a)da. (4.5) Multiplying both sides of (4.5) by q (a), and integrating, we obtain q (a)x 3(a)da = q (a) a This leads to the following characteristic equation: Let 1 = H (λ) := = q (a) a+ a q (a) K (σ)s 1(σ)e a a + σ (λ+µ(θ)+γ(θ))dθ dσda q (a)x 3(a)da. K (σ)s 1(σ)e a σ (λ+µ(θ)+γ (θ))dθ dσda. (4.6) a K (a)s 1(a) K (σ)s 1(σ)e a σ (λ+µ(θ)+γ (θ))dθ dσda Then H (λ) is a continuously differentiable function with a q (σ)e σ a (λ+µ(θ)+γ j (θ))dθ dσda. lim λ + H (λ) =, lim λ H (λ) = +. Furthermore, we have H (λ) <. It then follows that H (λ) is a decreasing function. So the equation (4.6) has a unique real root λ. Noting that H () = R 1, 11
12 we have λ < if R1 < 1, and λ > if R1 > 1. Let λ = ξ + ηi be an arbitrary complex root to equation (3.1). Then which implies that λ 1 = H (λ) = H (ξ + ηi) H (ξ), > ξ. Thus, all the roots of the equation (4.6) have negative real part if R 1 < 1. In addition, if R 1 > 1 we can easily see that 1 = H (λ) has a positive real root and E 1 is unstable. However, if R 1 < 1 the stability of E 1 is completely determined by the equations (4.4a) and (4.4b) under the condition that x 3 (a). Assumption (H5) implies that E 1 is locally stable if x 3 (a). Thus if R 1 < 1 then E 1 is locally stable. This completes the proof of Theorem Persistence of the pathogens In this section we present our main result: sufficient conditions for the persistence of two competing strains. One consequence of the persistence of the strains is that they coexist. Our principal result in this section can be stated as follows. Theorem 5.1. Assume R d > 1, R j > 1, j = 1, and (H1)-(H5) hold. If R1 > 1 and R1 > 1, then there exists a constant ε > such that every solution (s(a, t), i 1 (a, t), i (a, t)) with initial condition (s ( ), i 1 ( ), i ( )) Γ satisfies where lim inf t i 1 (a, t)da ε, lim inf t Γ := {(s ( ), i 1 ( ), i ( )) X + : i 1 (a)da >, i (a)da >, i (a, t)da ε, ã s (a)da > }. In order to prove Theorem 5.1, we need the following lemmas: Lemma 5.. Assume (H1) holds, then the system (.3) is dissipative. Proof. Since n(a, t) = s(a, t) + i 1 (a, t) + i (a, t), it follows from system (4.1) that n(a, t) satisfies the following differential equation: n(a, t) n(a, t) + t = µ(a)n(a, t) n(, t) = R d Φ(Q(t)) β(a)n(a, t)da, n(a, ) = n (a), 1 γ j (a)i j (a, t), j=1 (5.1)
13 where n (a) = s (a) + i j (a). Integrating the system (5.1) along the characteristic lines we get n(a, t) = j=1 RΦ(Q(t d a)) n (a t)e a a t (µ(θ)+ β(σ)n(σ, t a)dσe a (µ(θ)+ γ j (θ) i j (θ,θ+t a) n(θ,θ+t a) )dθ j=1, t > a, j=1 γ j (θ) i j (θ,θ+t a) n(θ,θ+t a) )dθ, t < a. Since Φ(x) is strictly decreasing, it follows that R d n(a, t) Φ(r min n(σ, t a)dσ)β max n(σ, t a)dσe a µ(θ)dθ, t > a, n (a t)e a a t µ(θ)dθ, t < a. Moreover, (H1) implies that n(a, t) Me a µ(θ)dθ, t > a, n (a t)e a a t µ(θ)dθ, t < a. (5.) It is easy to see that n(a, t) M for almost all a (, a + ) and t > a +. Thus the system (.3) is dissipative. Lemma 5.3. Assume (H1)-(H3) hold, then Γ is positively invariant for system (.3). Proof. Let (φ, ϕ, ψ) Γ, and (s(a, t), i 1 (a, t), i (a, t)) be the solution to system (.3) with the initial conditions s(, ) = φ, i 1 (, ) = ϕ, i (, ) = ψ. First, we prove that ã s(a, t)da > for all t >. To this aim, let us prove that the following assertion: Assertion: If s(, t) > for t (t 1, t ) then s(, t) > for t (t 1 + ã, t + ã). In fact, if t (t 1 +ã, t +ã), then ã (t t ) < ã (t t 1 ), and (ã (t t ), ã (t t 1 )) (ã, ã), {t (ã (t t ), ã (t t 1 ))} (t 1, t ). It follows from (.3) that we have s(, t) = RΦ(Q(t)) d β(a)n(a, t)da > R d Φ(M r(σ)dσ) = R d Φ(M >. r(σ)dσ) The proof of the assertion is completed. ã (t t1 ) ã (t t ) ã (t t1 ) ã (t t ) β(a)s(a, t)da β(a)s(, t a)e a (µ(θ)+ λ j (θ,θ+t a))dθ j=1 da 13
14 Because ã φ(a)da > and the way ã is defined, it follows that there exists a t (, ã) such that ã t β(a)φ(a t) >. Then we have s(, t) = RΦ(Q(t)) d β(a)s(a, t)da ã > RΦ(M d r(σ)dσ) β(a)φ(a t)e a a t (µ(θ)+ λ j (θ,θ+t a))dθ j=1 da >. t By the continuity of the function s(, t), we can conclude that there exists an interval (t 1, t ) such that t (t 1, t ) and s(, t) > for all t (t 1, t ). Iterating the above assertion, we have for any integer m s(, t) > t (t 1 + mã, t + mã). (5.3) We now are able to prove that ã s(a, t) > for all t >. If t [, t], then we have ã s(a, t) > ã t ã t >. φ(a t)e a a t (µ(θ)+ λ j (θ,θ+t a))dθ j=1 da φ(a t)e a a t (µ(θ)+ λ j (θ,θ+t a))dθ j=1 da If t (t 1, t + ã), then we have ã s(a, t) >. ã (t t1 ) (t t ) ã (t t1 ) (t t ) s(a, t)dt s(, t a)e a (µ(θ)+ λ j (θ,θ+t a))dθ j=1 da Similarly, if t (t 1 + ã, t + ã) then we have ã s(a, t)da >. Thus, iterating the above procedure yields that ã s(a, t)da > for all t. Second, let us show that i j (a, t)da >, j = 1, for all t >. To this aim, we divide the proof into three steps: Step 1. Define i jc (a) := i j (a, a + c) a [ ( c), a + ) for c > a +. Then di jc (a) da = λ j (a, a + c)s(a, a + c) (µ(a) + γ j (a))i jc (a) (µ(a) + γ j (a))i jc (a). 14 (5.4)
15 It follows from (5.4) that if i jc (a ) > we have i jc (a) >, i.e., i j (a, a + c) >, for all a > a. Step. Since ã s(a, )da >, there exists ā < ã + ã ã such that (ā+ ã ã ) ā s(a, )da >. In this step, we prove that di jc(a) da a= > for all c (ã ā, ã+ã ā). For ease of presentation, we assume that ā < ã, and if ā ã we can analyze in a similar way. Integrating the system (.3) along the characteristic lines yields a λ j (θ, θ + t a)s(θ, θ + t a)e a θ (µ(ξ)+γ j (ξ))dξ dθ, t > a, i j (a, t) = i j (a t)e a a t (µ(θ)+γ j (θ))dθ + a a t λ j (θ, θ + t a)s(θ, θ + t a)e a θ (µ(ξ)+γ j (ξ))dξ dθ, t < a. (5.5). Let c (ã ā, ã+ã ā), and consider two cases: ā+ ã ã ā Case 1) (ā+ ã ã ) ā c+ā+ ã ã c+ā i j (a)da >. Then it follows from (5.5) that i j (a, c)da > = c+ā+ ã ã c+ā ã ã ā+ ā >. i j (a)da > and ā+ ã ã ā i j (a)da = i j (a c)e a a c (µ(θ)+γ j (θ))dθ da i j (a)e a+c (µ(θ)+γ a j (θ))dθ da Consequently, we have (c + ā, c + ā + ã ã ) (ã, ã) and for all c (ã ā, ã+ã ā) di jc (a) da a= = λ j (, c)n(, c) = K j () K j () q j (σ)i j (σ, c)dσn(, σ) ã ã c+ā+ c+ā q j (σ)i j (σ, c)dσr d Φ(M K j ()q j min β min R d Φ(M r(σ))( >. Case ) (ā+ ã ã c+ā+ ã ã c+ā > ā ) i j (a, c)da c+ā+ ã ã c+ā ã ã c+ā+ c+ā i j (a)da =, i.e., (ā+ ã ã a a c ā ) ã ã c+ā+ c+ā ã ã c+ā+ r(σ)) β(σ)i j (σ, c)dσ c+ā i j (σ, c)dσ) φ(a)da >. It follows from (5.5) that λ j (θ, θ + c a)s(θ, θ + c a)e a θ (µ(ξ)+γ j (ξ))dξ dθda K j min q j min s(a, c) a a c 15 (5.6) i j (σ, θ + c a)dσe a θ (µ(ξ)+γ j (ξ))dξ dθda.
16 From the fact i j (σ, )dσ > and the continuity of solutions, we have f(a) := a a c i j (σ, θ + c a)dσe a θ (µ(ξ)+γ j (ξ))dξ dθ > for all a > c. Integrating the first equation of system (.3) along the characteristic lines, we get the form for a (c+ā, c + ā + ã ã ). Since (ā+ ã ã ā. Then we have c+ā+ ã ã c+ā i j (a, c)da K j min q j min c+ā+ >. c+ā s(a, c) = φ(a c)e a a c (µ(ξ)+ λ j (ξ,ξ+c a))dξ j=1 ã ã ) φ(a)da >, it follows that c+ā+ ã ã c+ā s(a, c)da > a s(a, c) i j (σ, θ + c a)dσe a θ (µ(ξ)+γ j (ξ))dξ dθda. a c Using a similar way as in the proof of the Case 1), we can conclude that (c + ā, c + ā+ ã ã ) (ã, ã) and i jc(a) da a= > for all c (ã ā, ã+ã ā). Step 3. In this step, we prove the following assertion: If di jc(a) da a= > for c (t 3, t 4 ), then di jc(a) da a= > for c (t 3 + ã, t 4 + ã). (5.7) In fact, if di jc(a) da a= > for c (t 3, t 4 ) then it follows from (5.4) that i j (a, c) > for all a ( (c t 4 ), (c t 3 ) a + ) and c (t 3, t 4 +a + ). Let c (t 3 +ã, t 4 +ã), then ã (c t 4 ) < ã (c t 3 ), and (ã (c t 4 ), ã (c t 3 )) (ã, ã), (ã (c t 4 ), ã (c t 3 )) ( (c t 4 ), (c t 3 ) a + ). Thus we have i j (a, c) > for all a (ã (c t 4 ), ã (c t 3 )). It follows from (5.4) and the fact that i jc () = that di jc (a) da a= = λ j (, c)s(, c) = K j () K j () K j () ã (c t3 ) q j (σ)i j (σ, c)dσs(, c) q j (σ)i j (σ, c)dσs(, c) ã (c t 4 ) q j (σ)i j (σ, c)dσr d Φ(M K j ()q j min β min R d Φ(M r(σ))( >. 16 ã (c t3 ) ã (c t 4 ) r(σ)) ã (c t3 ) ã (c t 4 ) i j (σ, c)dσ) β(σ)i j (σ, c)dσ (5.8)
17 Thus, the statement (5.7) is proved. It then follows from (5.4) that i j (a, c) > for all a ( (c (t 4 + ã)), (c (t 3 + ã )) a + ) and c (t 3 + ã, t 4 + ã + a + ). Since there exists ā < ã + ã ã such that di jc(a) da a= > for all c (ã ā, ã+ã ā), iterating (5.7) we have di jc (a) da a= > for all c (ã ā + nã, ã + ã ā + nã). (5.9) If there exists T such that i j (a, T )da =, then it is easy to see that i j (a, t) = is also a solution of (.3) for t > T. Thus we have di jc(a) da a= = for all t > T. This contradicts (5.9) due to the uniqueness of solutions to (.3). This contradiction implies that i j (a, t) > for all t >. This completes the proof Lemma 5.3. Proof. (Theorem 5.1): Define B := {(s( ), i 1 ( ), i ( )) X + : ã B 11 := {(s( ), i 1 ( ), i ( )) X + : ã s(a)da >, i j (a), j = 1, for almost all a (, a + )}; (s(a) + i 1 (a))da =, i (a) for almost all a (, a + )}; B 1 := {(s( ), i 1 ( ), i ( )) X + : ã (s(a) + i 1(a))da >, B 1 := {(s( ), i 1 ( ), i ( )) X + : ã i 1 (a)da >, i (a) for almost all a (, a + )}; (s(a) + i (a))da =, i 1 (a) for almost all a (, a + )}; B := {(s( ), i 1 ( ), i ( )) X + : ã (s(a) + i (a))da >, Γ := B B 11 B 1 B 1 B, X := Γ Γ. i (a)da >, i 1 (a) for almost all a (, a + )}; In order to show the theorem, it suffices to show that Γ repeals uniformly the solutions of Γ. We can easily see that B and B ij, i, j = 1, are positively invariant for system (.3). Thus Γ is positively invariant for system (.3). It follows from Lemma 5.3 that Γ is positively invariant for system (.3). Similar approach as in Proposition 3.16 by Webb [] can result in the fact that the 17
18 dynamical system is asymptotically smooth. From Theorem 4.1, Theorem 4.3 and the assumptions (H4) and (H5), we have the following conclusions: (1) E is a global attractor in B for system (.3); () E is a global attractor in B 11 B 1 for system (.3); (3) E 1 is a global attractor in B 1 for system (.3); (4) E is a global attractor in B for system (.3). Thus we have à := (s ( ),i 1 ( ),i ( )) Γ = {E, E, E 1, E }. ω((s ( ), i 1 ( ), i ( )) By the above conclusions, it follows that à is isolated and has an acyclic covering M = {E, E, E 1, E }. It follows from Lemma 5. that the system (.3) is dissipative and the orbit of any bounded set is bounded. By Theorem 4. in [4], in order to show that Γ repeals uniformly the solutions of Γ we only need to show that W s (E ) Γ =, W s (E ) Γ =, W s (E 1 ) Γ =, W s (E ) Γ = if R d > 1, R 1 > 1, R1 > 1. Since R d > 1, we can choose η 1 > small enough such that R d Φ(3η 1 r(σ)dσ)e η 1 j=1 a + K j (a)da q j (a)da > 1. (5.1) Assume that W s (E ) Γ. Then there exists a positive solution (ŝ(a, t), î 1 (a, t), î (a, t)) with the initial conditions ŝ(, ) = ψ 1, î 1 (, ) = ϕ 1, î (, ) = φ 1, (ψ 1, ϕ 1, φ 1 ) Γ such that (ŝ(a, t), î 1 (a, t), î (a, t)) E as t +. Thus there exists T 1 > such that for t > T 1 we have < ŝ(, t) < η 1, < î 1 (, t) < η 1, < î (a, t) < η 1, (ŝ(, T 1 ), î 1 (, T 1 ), î (, T 1 )) Γ, and ŝ(a, t) ŝ(a, t) + > (µ(a) + η 1 K j (a) q j (σ)dσ)ŝ(a, t), t j=1 ŝ(a, ) = ψ 1 (a), ŝ(, t) > R d Φ(3η 1 r(σ)dσ) β(a)ŝ(a, t)da. Consider the following auxiliary system (a, t) (a, t) + = (µ(a) + η 1 K j (a) q j (σ)dσ) (a, t), t j=1 (a, T ) = ŝ(a, T 1 ), (5.11) (, t) = R d Φ(3η 1 r(σ)dσ) β(a) (a, t)da. 18
19 By the comparison principle we have ŝ(, t) (, t) for all t > T 1. By the theory of linear systems, it is easy to see that (, t) + as t tends to infinity if (5.1) holds. This contradicts ŝ(, t) as t +. This contradiction implies that we have W s (E ) Γ =. Now we show that W s (E ) Γ =. Since R 1 > 1, R1 > 1 imply R j > 1, j = 1,, we can choose η > small enough such that a+ a+ K 1 (a)(n (a) η ) K (a)(n (a) η ) a a q 1 (σ)e σ a (µ(θ)+γ 1(θ))dθ dσda > 1, q (σ)e σ a (µ(θ)+γ (θ))dθ dσda > 1. Assume that W s (E ) Γ. Then there exists a positive solution ( s(a, t), ĩ 1 (a, t)), ĩ (a, t)) with initial conditions s(, ) = ψ, ĩ 1 (, ) = ϕ, ĩ (, ) = φ, (ψ, ϕ, φ ) Γ such that ( s(a, t), ĩ 1 (a, t), ĩ (a, t)) E as t +. Thus there exists T > such that for t > T we have n ( ) η < s(, t) < n ( ) + η, < ĩ 1 (, t) < η, < ĩ (a, t) < η, for ( s(, T ), ĩ 1 (, T ), ĩ (a, T )) Γ, and ĩ 1 (a, t) + ĩ 1 (a, t) a + > K 1 (a) q 1 (σ)ĩ 1 (σ, t)dσ(n (a) η ) (µ(a) + γ 1 (a))ĩ 1 (a, t), t ĩ (a, t) t + ĩ (a, t) ĩ 1 (a, ) = ϕ (a), ĩ (a, ) = φ (a), > K (a) q (σ)ĩ (σ, t)dσ(n (a) η ) (µ(a) + γ (a))ĩ (a, t), ĩ 1 (, t) = ĩ (, t) =. Consider the auxiliary system l 1 (a, t) t l (a, t) t + l 1(a, t) + l (a, t) l 1 (a, T ) = ĩ 1 (a, T ), = K 1 (a) = K (a) q 1 (σ)l 1 (σ, t)dσ(n (a) η ) (µ(a) + γ 1 (a))l 1 (a, t), q (σ)l (σ, t)dσ(n (a) η ) (µ(a) + γ (a))l (a, t), l (a, T ) = ĩ (a, T ), l 1 (, t) =, l (, t) =. (5.1) 19
20 By the comparison principle, we can prove that ĩ j (, t) l j (, t), j = 1, for all t > T. From the theory of linear systems it is easy to see that l j (, t) +, j = 1, as t tends to infinity if a+ K j (a)(n (a) η ) a q j (σ)e σ a (µ(θ)+γ j (θ))dθ dσda > 1, j = 1,. This contradicts ĩ j (, t), j = 1, as t +. This contradiction implies that we have W s (E ) Γ =. In the following, we show that W s (E 1 ) Γ =, W s (E ) Γ =. Here we only show that W s (E 1 ) Γ = if R 1 > 1. If R1 > 1, W s (E ) Γ = can be proved in a similar way. Since R 1 > 1, we can choose η 3 > small enough such that a+ K (a)(s 1 (a) η 3) q (σ)e σ a (µ(θ)+γ (θ))dθ dσda > 1. Assume that W s (E 1 ) Γ. Then there exists a positive solution (š(a, t), ǐ 1 (a, t), ǐ (a, t)) with initial conditions š(, ) = ψ 3, ǐ 1 (, ) = ϕ 3, ǐ (, ) = φ 3, (ψ 3, ϕ 3, φ 3 ) Γ such that (š(a, t), ǐ 1 (a, t), ǐ (a, t)) E 1 as t +. Thus there exists a T 3 > such that for t > T 3 we have s 1 ( ) η 3 < š(, t) < s 1 ( ) + η 3, i 1 ( ) η 3 < ǐ 1 (, t) < i 1 ( ) + η 3, < ǐ (a, t) < η 3, and ǐ (a, t) t + ǐ (a, t) ǐ (a, ) = ϕ (a), ǐ (, t) =. (š(, T 3 ), ǐ 1 (, T 3 ), ǐ (, T 3 )) Γ, > K (a) q (σ)ǐ (σ, t)dσ(s 1(a) η 3 ) (µ(a) + γ (a))ǐ (a, t), Consider the auxiliary system l (a, t) t + l (a, t) l (a, T 3 ) = ǐ (a, T 3 ), l (, t) =. = K (a) q (σ)l (σ, t)dσ(s 1 (a) η 3) (µ(a) + γ (a))l (a, t), (5.13) By the comparison principle, we can prove that ǐ (, t) l (, t) for all t > T 3. From the theory of linear systems it is easy to see that l (a, t)da + as t tends to infinity if a+ K (a)(s 1 (a) η 3) q (σ)e σ a (µ(θ)+γ(θ))dθ dσda > 1. This contradicts ǐ (, t) as t +. This contradiction implies that we have W s (E 1 ) Γ =. Since W s (E ) Γ =, W s (E ) Γ =, W s (E j ) Γ =, j = 1,, and {E, E, E 1, E } are acyclic in Γ, by Theorem 4. in [4] we are able to conclude that the system (.3) is uniformly persistent with respect to (Γ, Γ). This completes the proof of Theorem 5.1.
21 6 Numerical results In this section, we present some numerical results that support and extend the analytical results. Backward Euler and the linearized finite difference method are used to discretize the PDEs, and the integral is evaluated using trapezoidal rule. For illustration purposes, we choose a + =, R d =. The functions µ(a), β(a), r(a), γ j(a), j = 1, are assumed to be constant. These parameters are chosen as follows: µ(a).6, β(a).6, r(a) 1, γ j (a), q j (a) 1, j = 1,, a [, ]. The functions Φ(x), K 1 (a), K (a) are chosen to have the following forms: Φ(x) = x ; K 1 (a) = m 1 1+k 1 a ; K (a) = m + k a. The simulation results are shown in Figure 1. Figure 1 (a) is for the case k 1 =, k =, m 1 =.1, m =.6. In this case system (.3) does not depend on the age structure, and it follows from Theorem 3.1 that the strain with the larger reproduction number, i.e., strain 1, uniformly persists, and the other strain dies out. Figure 1 (b) is for the case m 1 =.1, k 1 = 1., m =.6, k =.6. In this case, we can obtain that R 1 = 1.91, R 1 = Since both invasion numbers are larger than one, we expect persistence of the strains. Indeed, as Figure 1 (b) demonstrates the two competing strains coexist in a unique positive equilibrium which as simulations seem to suggest seems globally attracting. Figure 1 (c) is for the case m 1 =.1, k 1 = 1., m =.3, k =.6. In the case, the system (.3) has four boundary equilibria E, E, E 1, E. Straight forward computation yields that R 1 =.68, R1 = It follows from Theorem 4.3 that E is locally stable under assumption (H4). Figure 1 (c) is showing that E is indeed asymptotically stable. Figure 1 (d) is for the case m 1 =.1, k 1 = 1., m =.9, k =.6. In this case, we have that R 1 = , R1 =.9431 and the system (.3) also has four boundary equilibria. In Figure 1 (d), we can easily see that the boundary equilibrium E 1 is asymptotically stable. This simulation results have either confirm or extended our analytical results. 7 Discussion In this paper, we have studied an age-structured epidemic model with two competing strains. The main focus of the paper is the proof of persistence of the two strains when both invasion numbers are greater than one. In our case the coexistence of the strains is induced by the host agestructure. Although the presence of disease-induced mortality into the age-structured model makes 1
22 6 5 5 I 1 I 4 I 1 I t (a) t (b) I 1 I I 1 I t (c) t (d) Figure 1: Numerical solutions of the system (.3). I 1 and I represent the number of the infected cases with strain one and two, respectively. The values of of m j, k j, j = 1, used in the four plots are: (a) m 1 =.1, m =.6, k 1 = k = (age independent case); (b) m 1 =.1, m =.6, k 1 = 1., k =.6; (c) m 1 =.1, m =.3, k 1 = 1., k =.6; and (d) m 1 =.1, m =.9, k 1 = 1., k =.6.
23 the analysis of the resulting system very difficulty, we are able to carry out systematic analysis of the model and establish rigorously uniform persistence of both strains. The basic reproduction numbers R j for both strains are defined in section 3, and Theorem 4. shows the uniform persistence of the single strain j in the case when R j > 1. We have also obtained the explicit expression of the invasion reproduction numbers R j k, j, k = 1,, j k for both strains. Local stabilities of the boundary equilibria are discussed in Theorem 4.3. By the theory of uniform persistence of infinite dimensional dynamical systems the coexistence of two competing strains is rigorously proved under the conditions that both invasion reproduction numbers are lager than one. However, results for the corresponding age-independent model are summarized in section 3 and show that the two competing strains can not coexist. This indicates that age structure leads to the coexistence of the strains. In section 6 numerical simulations are further conducted to confirm and extend the analytical results. Finally, there are still many interesting and challenging mathematical questions that remain open for system (.3). For example, we could not present results on the global dynamics of system (.3). It seems very difficult to analyze the global dynamics for that system. However, if we neglect the disease-induced death rates, then system (.3) can be reduced to a competitive system. We may use techniques from monotone theory developed in [5, 3] to provide the global behavior of the reduced model. Numerical simulations suggest that if R 1 > 1, R > 1 and the invasion numbers are greater than one, the reduced system (.3) has a unique positive equilibrium which is globally asymptotically stable. We conclude the discussion by formulating the following conjecture: Conjecture: Assume R 1 > 1, R > 1 and the invasion reproduction numbers are greater than one. Then system (.3) with γ 1 (a) = γ (a) = has a unique coexistence equilibrium which is globally asymptotically stable. Acknowledgements Zhipeng Qiu greatly appreciates Prof. Iannelli and Prof. Pugliese for their encouragement and many good suggestions. This work is supported by Istituto Nazionale di Alta Matematica (INdAM)- Rome and Italian Ministry of University and Research (MUR) within the China-Italy Executive Cooperation Programme of the Italian Ministry of Foreign Affairs (MAE), the NSFC grants (No , , ), and the US NSF under grant DMS
24 Appendix: Proof of Theorem 4.1 Proof. Using a similar approach we can easily prove the second conclusion of Theorem 4.1 ). Here we only present the proofs for Theorem 4.1 1) and the first conclusion of Theorem 4.1 ). Let ñ(a, t) be the solution of the following system ñ(a, t) t + ñ(a, t) ñ(a, ) = n (a), = µ(a)ñ(a, t), ñ(, t) = R d Φ() β(a)ñ(a, t)da. We claim that n(a, t) ñ(a, t) for all t and almost all a (, a + ). In order to prove the claim, let us integrate the system (.1) and (7.1) along the characteristic lines. Then we get and n(a, t) = ñ(a, t) = n(, t a)e a µ(θ)dθ, t > a, n (a t)e a a t µ(θ)dθ, t < a. ñ(, t a)e a µ(θ)dθ, t > a, n (a t)e a a t µ(θ)dθ, t < a. We now show that n(, t) ñ(, t). Suppose the contrary, that is the inequality does not hold. Then we can define t such that t := inf{t > : n(, t) > ñ(, t)} since n(, ) < ñ(, ) and n(, t), ñ(, t) are both continuous functions with respect to t. Consequently, we have R d Φ(Q(t )) β(a)n(a, t )da = R Φ() β(a)ñ(a, t )da. Since Φ(Q(t )) Φ() it follows that there exist ā (, t ) such that n(ā, t ) ñ(ā, t ). It follows from (7.) and (7.3) that we have n(, t ā) ñ(, t ā). This contradicts the definition of t. This contradiction implies that n(, t) ñ(, t). From (7.) and (7.3) we get n(a, t) ñ(a, t) for all t > a >. When a > t >, we can easily see that n(a, t) = ñ(a, t). Thus n(a, t) ñ(a, t) for all t and almost all a (, a + ). If R d < 1, it is easy to see that lim ñ(, t, n (a)) = for all n (a) L 1 +(, a + ). From the t + above claim it also follows that we have lim n(, t, n (a)) = for each n ( ) L 1 t + + (, a+ ). This complete the proof of Theorem 4.1 1). (7.1) (7.) (7.3) 4
25 Assume R d > 1 and n (a) L 1 +(, a + ). If ã n (a)da =, it then follows from Theorem 5.1 of paper [5] that we have Similarly, we have This completes the proof of Theorem 4.1. lim ñ(, t, n (a)) =. t + lim t + n(, t, n (a)) =. References [1] F. Carrat, J. Luong, H. Lao, A. Salle, C. Lajaunie, H.Wackernagel, A small-world-like model for comparing interventions aimed at preventing and controlling influenza pandemics. BMC Med., 6, 4: 6. doi:1.1186/ [] S. N. Busenberg, M. Iannelli, H. R. Thieme, Global behavior of an age-structured epdiemic model, SIAM J. Math. Anal., 1991, : [3] H. Inaba, Mathematical analysis of an age-structured SIR epidemic model with vertical transmission, Dis. Con. Dyn. Sys. B, 6,6: [4] P. Zhang, Z. Feng, F. Milner, A schistosomiasis model with an age-structure in human hosts and its application to treatment strategies, Math.Biosci.,7, 5: [5] C. Castillo-Chavez, Z. Feng, Global stability of an age-structure model for TB and its applications to optimal vaccination strategies, Math. Biosci., 1998, 151: [6] J. S. Oxford, Influenza A pandemics of the th century with special reference to 1918: virology, pathology and epidemiology, Rev. Med. Vir.,, 1: [7] Centers for Disease Control and Prevention, West Nile Virus: Fact Sheet, cdc.gov/ncidod/dvbid/westnile/wnv factsheet.htm. [8] Z. Feng, J. X. Velasco-Hernandes, Competitive exclusion in a vector-host model for the dengue fever, J. Math. Biol., 1997, 35: [9] M. Martcheva, M. Iannelli, X. Z. Li, Subthreshold coexistence of strains: the impact of vaccination and mutation, Math. Biosci. Eng., 7, 4: [1] S. A. Levin, Community equilibria and stability, and an extension of the competitive exclusion principle, Am. Naturalist, 197, 14:
26 [11] H. J. Bremermann, H. R. Thieme, A competitive exclusion principle for pathogen virulence, J. Math. Biol., 1989, 7: [1] M. Iannelli, M. Martcheva, X. Z. Li, Strain replacement in an epidemic model with superinfection and perfect vaccination, Math. Biosci., 5, 195:3-46. [13] M. Martcheva, S.S. Pilyugin, The role of coinfection in multidisease dynamics, SIAM J. Appl. Math., 6, 66: [14] M. Martcheva, S.S. Pilyugin, R.D. Holt, Subthreshold and superthreshold coexistence of pathogen variants: The impact of host age-structure, Math. Biosci. 7, 7 (1): [15] M. Martcheva, H. Thieme, Progression age enhanced backward bifurcation in an epidemic model with super-infection, J. Math. Biol., 3, 46 (5): [16] R. May, M. Nowak, Coinfection and the evolution of parasite virulence, Proc. R. Soc. B, 1995, 61: [17] C. Castillo-Chavez, H. Hethcote, V. Andreasen, S. Levin, W. M. Liu, Epidemiological models with age structure, proportionate mixing, and cross-immunity, J. Math. Biol., 1989, 7: [18] V. Andreasen, A. Pugliese, Viral coexistence can be induced by density dependent host mortality, J. Theoret. Biol., 1995, 177: [19] M.E. Gurtin, R.C. MacCamy, Nonlinear age-dependent population dynamics, Archiv. Rat. Mech. Anal., 1974, 54:81-3. [] G. Webb, Theory of Nonlinear Age-Dependent Population Dynamics, Marcel Dekker, New York, [1] D. Breda, M. Iannelli, S. Maset, R. Vermiglio, Stability analysis of the Gurtin-MacCamy model, SIAM J. Numer. Anal., 8, 46: [] M. Iannelli, Mathematical theory of age-structured population dynamics, Applied Mathematics Monograph C.N.R. 7, Giardini Ed., Pisa [3] D. Breda, D. Visetti, Existence, uniqueness and multiplicity of endemic states for an agestructured S-I epidemic model, preprint. [4] J. K. Hale, P Waltmam, Persistence in infinite-dimensional systems, SIAM J. Math. Anal., 1989, :
27 [5] Z. Feng, W. Huang, C. Castillo-Chavez, Global behavior of a multi-group SIS epidemic model with age structure, J. Diff. Equs., 5, 18: [6] F. A. Milner, A. Pugliese, Periodic solutions: a robust numerical method for an SIR model of epidemics, J. Math. Biol., 1999, 39: [7] X. Z. Li, X. C. Duan, M. Ghosh, X. Y. Ruan, Pathogen coexistence induced by saturating contact rates, Nonl. Anal.: R.W.A., 9, 1: [8] P. van den Driessche, J. Watmough, Reproduction numbers and sub-threshold endemic equilibria for compartment models of disease transmission, Math. Biosci.,, 18: [9] H.R. Thieme, Persistence under relaxed point-dissipativity with an application to an epidemic model, SIAM J. Math. Anal., 1993, 4: [3] H. L. Smith, Monotone dynamical systems: an introduction to theory of competitive and cooperative systems, Math. Surveys Monogr., 41, AMS, Providence, RI,
SUBTHRESHOLD AND SUPERTHRESHOLD COEXISTENCE OF PATHOGEN VARIANTS: THE IMPACT OF HOST AGE-STRUCTURE
SUBTHRESHOLD AND SUPERTHRESHOLD COEXISTENCE OF PATHOGEN VARIANTS: THE IMPACT OF HOST AGE-STRUCTURE MAIA MARTCHEVA, SERGEI S. PILYUGIN, AND ROBERT D. HOLT Abstract. It is well known that in the most general
More information6. Age structure. for a, t IR +, subject to the boundary condition. (6.3) p(0; t) = and to the initial condition
6. Age structure In this section we introduce a dependence of the force of infection upon the chronological age of individuals participating in the epidemic. Age has been recognized as an important factor
More informationAN AGE-STRUCTURED TWO-STRAIN EPIDEMIC MODEL WITH SUPER-INFECTION. Xue-Zhi Li and Ji-Xuan Liu. Maia Martcheva
MATHEMATICAL BIOSCIENCES AND ENGINEERING Volume xx, Number xx, xx 2xx doi:.3934/mbe.29.xx.xx pp. xx AN AGE-STRUCTURED TWO-STRAIN EPIDEMIC MODEL WITH SUPER-INFECTION Xue-Zhi Li and Ji-Xuan Liu Department
More information1. Introduction: time-continuous linear population dynamics. Differential equation and Integral equation. Semigroup approach.
Intensive Programme Course: Lecturer: Dates and place: Mathematical Models in Life and Social Sciences Structured Population Dynamics in ecology and epidemiology Jordi Ripoll (Universitat de Girona, Spain)
More informationMathematical models on Malaria with multiple strains of pathogens
Mathematical models on Malaria with multiple strains of pathogens Yanyu Xiao Department of Mathematics University of Miami CTW: From Within Host Dynamics to the Epidemiology of Infectious Disease MBI,
More informationApparent paradoxes in disease models with horizontal and vertical transmission
Apparent paradoxes in disease models with horizontal and vertical transmission Thanate Dhirasakdanon, Stanley H.Faeth, Karl P. Hadeler*, Horst R. Thieme School of Life Sciences School of Mathematical and
More informationGLOBAL DYNAMICS OF A MATHEMATICAL MODEL OF TUBERCULOSIS
CANADIAN APPIED MATHEMATICS QUARTERY Volume 13, Number 4, Winter 2005 GOBA DYNAMICS OF A MATHEMATICA MODE OF TUBERCUOSIS HONGBIN GUO ABSTRACT. Mathematical analysis is carried out for a mathematical model
More informationGLOBAL STABILITY OF SIR MODELS WITH NONLINEAR INCIDENCE AND DISCONTINUOUS TREATMENT
Electronic Journal of Differential Equations, Vol. 2015 (2015), No. 304, pp. 1 8. SSN: 1072-6691. URL: http://ejde.math.txstate.edu or http://ejde.math.unt.edu ftp ejde.math.txstate.edu GLOBAL STABLTY
More informationGlobal Analysis of an Epidemic Model with Nonmonotone Incidence Rate
Global Analysis of an Epidemic Model with Nonmonotone Incidence Rate Dongmei Xiao Department of Mathematics, Shanghai Jiaotong University, Shanghai 00030, China E-mail: xiaodm@sjtu.edu.cn and Shigui Ruan
More informationIntroduction to SEIR Models
Department of Epidemiology and Public Health Health Systems Research and Dynamical Modelling Unit Introduction to SEIR Models Nakul Chitnis Workshop on Mathematical Models of Climate Variability, Environmental
More informationGlobal Stability of SEIRS Models in Epidemiology
Global Stability of SRS Models in pidemiology M. Y. Li, J. S. Muldowney, and P. van den Driessche Department of Mathematics and Statistics Mississippi State University, Mississippi State, MS 39762 Department
More informationReproduction numbers and sub-threshold endemic equilibria for compartmental models of disease transmission
Reproduction numbers and sub-threshold endemic equilibria for compartmental models of disease transmission P. van den Driessche a,1 and James Watmough b,2, a Department of Mathematics and Statistics, University
More informationThursday. Threshold and Sensitivity Analysis
Thursday Threshold and Sensitivity Analysis SIR Model without Demography ds dt di dt dr dt = βsi (2.1) = βsi γi (2.2) = γi (2.3) With initial conditions S(0) > 0, I(0) > 0, and R(0) = 0. This model can
More informationFarkas JZ (2004) Stability conditions for the non-linear McKendrick equations, Applied Mathematics and Computation, 156 (3), pp
Farkas JZ (24) Stability conditions for the non-linear McKendrick equations, Applied Mathematics and Computation, 156 (3), pp. 771-777. This is the peer reviewed version of this article NOTICE: this is
More informationLa complessa dinamica del modello di Gurtin e MacCamy
IASI, Roma, January 26, 29 p. 1/99 La complessa dinamica del modello di Gurtin e MacCamy Mimmo Iannelli Università di Trento IASI, Roma, January 26, 29 p. 2/99 Outline of the talk A chapter from the theory
More informationBehavior Stability in two SIR-Style. Models for HIV
Int. Journal of Math. Analysis, Vol. 4, 2010, no. 9, 427-434 Behavior Stability in two SIR-Style Models for HIV S. Seddighi Chaharborj 2,1, M. R. Abu Bakar 2, I. Fudziah 2 I. Noor Akma 2, A. H. Malik 2,
More informationGlobal Analysis of an SEIRS Model with Saturating Contact Rate 1
Applied Mathematical Sciences, Vol. 6, 2012, no. 80, 3991-4003 Global Analysis of an SEIRS Model with Saturating Contact Rate 1 Shulin Sun a, Cuihua Guo b, and Chengmin Li a a School of Mathematics and
More informationPermanence Implies the Existence of Interior Periodic Solutions for FDEs
International Journal of Qualitative Theory of Differential Equations and Applications Vol. 2, No. 1 (2008), pp. 125 137 Permanence Implies the Existence of Interior Periodic Solutions for FDEs Xiao-Qiang
More informationGlobal analysis of multi-strains SIS, SIR and MSIR epidemic models
Global analysis of multi-strains I, IR and MIR epidemic models Derdei Bichara, Abderrahman Iggidr, Gauthier allet To cite this version: Derdei Bichara, Abderrahman Iggidr, Gauthier allet. Global analysis
More informationEpidemics in Two Competing Species
Epidemics in Two Competing Species Litao Han 1 School of Information, Renmin University of China, Beijing, 100872 P. R. China Andrea Pugliese 2 Department of Mathematics, University of Trento, Trento,
More informationAge-time continuous Galerkin methods for a model of population dynamics
Journal of Computational and Applied Mathematics 223 (29) 659 671 www.elsevier.com/locate/cam Age-time continuous Galerkin methods for a model of population dynamics Mi-Young Kim, Tsendauysh Selenge Department
More informationAvailable online at Commun. Math. Biol. Neurosci. 2015, 2015:29 ISSN:
Available online at http://scik.org Commun. Math. Biol. Neurosci. 215, 215:29 ISSN: 252-2541 AGE-STRUCTURED MATHEMATICAL MODEL FOR HIV/AIDS IN A TWO-DIMENSIONAL HETEROGENEOUS POPULATION PRATIBHA RANI 1,
More informationUNIFORM WEAK IMPLIES UNIFORM STRONG PERSISTENCE FOR NON-AUTONOMOUS SEMIFLOWS
PROCEEDINGS OF THE AMERICAN MATHEMATICAL SOCIETY Volume 127, Number 8, Pages 2395 2403 S 0002-9939(99)05034-0 Article electronically published on April 15, 1999 UNIFORM WEAK IMPLIES UNIFORM STRONG PERSISTENCE
More informationGLOBAL STABILITY OF THE ENDEMIC EQUILIBRIUM OF A TUBERCULOSIS MODEL WITH IMMIGRATION AND TREATMENT
CANADIAN APPLIED MATHEMATICS QUARTERLY Volume 19, Number 1, Spring 2011 GLOBAL STABILITY OF THE ENDEMIC EQUILIBRIUM OF A TUBERCULOSIS MODEL WITH IMMIGRATION AND TREATMENT HONGBIN GUO AND MICHAEL Y. LI
More informationTransmission Dynamics of an Influenza Model with Vaccination and Antiviral Treatment
Bulletin of Mathematical Biology (2010) 72: 1 33 DOI 10.1007/s11538-009-9435-5 ORIGINAL ARTICLE Transmission Dynamics of an Influenza Model with Vaccination and Antiviral Treatment Zhipeng Qiu a,, Zhilan
More informationThe effect of population dispersal on the spread of a disease
J. Math. Anal. Appl. 308 (2005) 343 364 www.elsevier.com/locate/jmaa The effect of population dispersal on the spread of a disease Yu Jin, Wendi Wang Department of Mathematics, Southwest China Normal University,
More informationSTABILITY AND PERSISTENCE IN A MODEL FOR BLUETONGUE DYNAMICS
STABILITY AND PERSISTENCE IN A MODEL FOR BLUETONGUE DYNAMICS STEPHEN A. GOURLEY, HORST R. THIEME, AND P. VAN DEN DRIESSCHE Abstract. A model for the time evolution of bluetongue, a viral disease in sheep
More informationQualitative Analysis of a Discrete SIR Epidemic Model
ISSN (e): 2250 3005 Volume, 05 Issue, 03 March 2015 International Journal of Computational Engineering Research (IJCER) Qualitative Analysis of a Discrete SIR Epidemic Model A. George Maria Selvam 1, D.
More informationDelay SIR Model with Nonlinear Incident Rate and Varying Total Population
Delay SIR Model with Nonlinear Incident Rate Varying Total Population Rujira Ouncharoen, Salinthip Daengkongkho, Thongchai Dumrongpokaphan, Yongwimon Lenbury Abstract Recently, models describing the behavior
More informationGLOBAL DYNAMICS OF A TIME-DELAYED DENGUE TRANSMISSION MODEL
CANADIAN APPLIED MATHEMATICS QUARTERLY Volume 2, Number 1, Spring 212 GLOBAL DYNAMICS OF A TIME-DELAYED DENGUE TRANSMISSION MODEL Dedicated to Herb Freedman on the occasion of his 7th birthday ZHEN WANG
More informationStability of SEIR Model of Infectious Diseases with Human Immunity
Global Journal of Pure and Applied Mathematics. ISSN 0973-1768 Volume 13, Number 6 (2017), pp. 1811 1819 Research India Publications http://www.ripublication.com/gjpam.htm Stability of SEIR Model of Infectious
More informationSTABILITY ANALYSIS OF A GENERAL SIR EPIDEMIC MODEL
VFAST Transactions on Mathematics http://vfast.org/index.php/vtm@ 2013 ISSN: 2309-0022 Volume 1, Number 1, May-June, 2013 pp. 16 20 STABILITY ANALYSIS OF A GENERAL SIR EPIDEMIC MODEL Roman Ullah 1, Gul
More informationThe dynamics of disease transmission in a Prey Predator System with harvesting of prey
ISSN: 78 Volume, Issue, April The dynamics of disease transmission in a Prey Predator System with harvesting of prey, Kul Bhushan Agnihotri* Department of Applied Sciences and Humanties Shaheed Bhagat
More informationHETEROGENEOUS MIXING IN EPIDEMIC MODELS
CANADIAN APPLIED MATHEMATICS QUARTERLY Volume 2, Number 1, Spring 212 HETEROGENEOUS MIXING IN EPIDEMIC MODELS FRED BRAUER ABSTRACT. We extend the relation between the basic reproduction number and the
More informationCan multiple species of Malaria co-persist in a region? Dynamics of multiple malaria species
Can multiple species of Malaria co-persist in a region? Dynamics of multiple malaria species Xingfu Zou Department of Applied Mathematics University of Western Ontario London, Ontario, Canada (Joint work
More informationResearch Article A Delayed Epidemic Model with Pulse Vaccination
Hindawi Publishing Corporation Discrete Dynamics in Nature and Society Volume 2008, Article ID 746951, 12 pages doi:10.1155/2008/746951 Research Article A Delayed Epidemic Model with Pulse Vaccination
More informationDynamics of Disease Spread. in a Predator-Prey System
Advanced Studies in Biology, vol. 6, 2014, no. 4, 169-179 HIKARI Ltd, www.m-hikari.com http://dx.doi.org/10.12988/asb.2014.4845 Dynamics of Disease Spread in a Predator-Prey System Asrul Sani 1, Edi Cahyono
More informationDynamics of a competing two-strain SIS epidemic model on complex networks with a saturating incidence rate
Dynamics of a competing two-strain SIS epidemic model on complex networks with a saturating incidence rate Junyuan Yang a,b, Chun-Hsien Li c, a Complex Systems Research Centre, Shanxi University, Taiyuan,
More information(mathematical epidemiology)
1. 30 (mathematical epidemiology) 2. 1927 10) * Anderson and May 1), Diekmann and Heesterbeek 3) 7) 14) NO. 538, APRIL 2008 1 S(t), I(t), R(t) (susceptibles ) (infectives ) (recovered/removed = βs(t)i(t)
More informationMathematical Modeling and Analysis of Infectious Disease Dynamics
Mathematical Modeling and Analysis of Infectious Disease Dynamics V. A. Bokil Department of Mathematics Oregon State University Corvallis, OR MTH 323: Mathematical Modeling May 22, 2017 V. A. Bokil (OSU-Math)
More informationTHE MATHEMATICAL MODELING OF EPIDEMICS. by Mimmo Iannelli Mathematics Department University of Trento. Lecture 4: Epidemics and demography.
THE MATHEMATICAL MODELING OF EPIDEMICS by Mimmo Iannelli Mathematics Department University of Trento Lecture 4: Epidemics and demography. THE MATHEMATICAL MODELING OF EPIDEMICS Lecture 4: Epidemics and
More informationA Time Since Recovery Model with Varying Rates of Loss of Immunity
Bull Math Biol (212) 74:281 2819 DOI 1.17/s11538-12-978-7 ORIGINAL ARTICLE A Time Since Recovery Model with Varying Rates of Loss of Immunity Subhra Bhattacharya Frederick R. Adler Received: 7 May 212
More informationModeling the Spread of Epidemic Cholera: an Age-Structured Model
Modeling the Spread of Epidemic Cholera: an Age-Structured Model Alen Agheksanterian Matthias K. Gobbert November 20, 2007 Abstract Occasional outbreaks of cholera epidemics across the world demonstrate
More informationStability of a Numerical Discretisation Scheme for the SIS Epidemic Model with a Delay
Stability of a Numerical Discretisation Scheme for the SIS Epidemic Model with a Delay Ekkachai Kunnawuttipreechachan Abstract This paper deals with stability properties of the discrete numerical scheme
More informationReceived 30 January 2003 and in revised form 8 February 2004
ANALYSIS OF AN AGE-DEPENDENT SI EPIDEMIC MODEL WITH DISEASE-INDUCED MORTALITY AND PROPORTIONATE MIXING ASSUMPTION: THE CASE OF VERTICALLY TRANSMITTED DISEASES M. EL-DOMA Received 3 January 23 and in revised
More informationMathematical Model of Tuberculosis Spread within Two Groups of Infected Population
Applied Mathematical Sciences, Vol. 10, 2016, no. 43, 2131-2140 HIKARI Ltd, www.m-hikari.com http://dx.doi.org/10.12988/ams.2016.63130 Mathematical Model of Tuberculosis Spread within Two Groups of Infected
More informationAccepted Manuscript. Backward Bifurcations in Dengue Transmission Dynamics. S.M. Garba, A.B. Gumel, M.R. Abu Bakar
Accepted Manuscript Backward Bifurcations in Dengue Transmission Dynamics S.M. Garba, A.B. Gumel, M.R. Abu Bakar PII: S0025-5564(08)00073-4 DOI: 10.1016/j.mbs.2008.05.002 Reference: MBS 6860 To appear
More informationModelling the spread of bacterial infectious disease with environmental effect in a logistically growing human population
Nonlinear Analysis: Real World Applications 7 2006) 341 363 www.elsevier.com/locate/na Modelling the spread of bacterial infectious disease with environmental effect in a logistically growing human population
More informationA GRAPH-THEORETIC APPROACH TO THE METHOD OF GLOBAL LYAPUNOV FUNCTIONS
PROCEEDINGS OF THE AMERICAN MATHEMATICAL SOCIETY Volume 136, Number 8, August 2008, Pages 2793 2802 S 0002-993908)09341-6 Article electronically published on March 27, 2008 A GRAPH-THEORETIC APPROACH TO
More informationA REMARK ON THE GLOBAL DYNAMICS OF COMPETITIVE SYSTEMS ON ORDERED BANACH SPACES
PROCEEDINGS OF THE AMERICAN MATHEMATICAL SOCIETY Volume 00, Number 0, Pages 000 000 S 0002-9939(XX)0000-0 A REMARK ON THE GLOBAL DYNAMICS OF COMPETITIVE SYSTEMS ON ORDERED BANACH SPACES KING-YEUNG LAM
More informationGLOBAL DYNAMICS OF A TWO-STRAIN DISEASE MODEL WITH LATENCY AND SATURATING INCIDENCE RATE
CANADIAN APPLIED MATHEMATIC QUARTERLY Volume 2, Number 1, pring 212 GLOBAL DYNAMIC OF A TWO-TRAIN DIEAE MODEL WITH LATENCY AND ATURATING INCIDENCE RATE Dedicated to Professor H.I. Freedman s 7th birthday.
More informationLinearized stability of structured population dynamical models
Linearized stability of structured population dynamical models Selected results from the thesis József Zoltán Farkas Supervisor: Prof.Dr. Miklós Farkas Budapest University of Technology Department of Differential
More informationGLOBAL STABILITY OF A VACCINATION MODEL WITH IMMIGRATION
Electronic Journal of Differential Equations, Vol. 2015 (2015), No. 92, pp. 1 10. SSN: 1072-6691. URL: http://ejde.math.txstate.edu or http://ejde.math.unt.edu ftp ejde.math.txstate.edu GLOBAL STABLTY
More informationMODELING AND ANALYSIS OF THE SPREAD OF CARRIER DEPENDENT INFECTIOUS DISEASES WITH ENVIRONMENTAL EFFECTS
Journal of Biological Systems, Vol. 11, No. 3 2003 325 335 c World Scientific Publishing Company MODELING AND ANALYSIS OF THE SPREAD OF CARRIER DEPENDENT INFECTIOUS DISEASES WITH ENVIRONMENTAL EFFECTS
More informationc 2006 Society for Industrial and Applied Mathematics
SIAM J. APPL. MATH. Vol. 66, No. 5, pp. 1563 1587 c 2006 Society for Industrial and Applied Mathematics DISCRETE-TIME SIS EPIDEMIC MODEL IN A SEASONAL ENVIRONMENT JOHN E. FRANKE AND ABDUL-AZIZ YAKUBU Abstract.
More informationStability and bifurcation analysis of epidemic models with saturated incidence rates: an application to a nonmonotone incidence rate
Published in Mathematical Biosciences and Engineering 4 785-85 DOI:.3934/mbe.4..785 Stability and bifurcation analysis of epidemic models with saturated incidence rates: an application to a nonmonotone
More informationGlobal Properties for Virus Dynamics Model with Beddington-DeAngelis Functional Response
Global Properties for Virus Dynamics Model with Beddington-DeAngelis Functional Response Gang Huang 1,2, Wanbiao Ma 2, Yasuhiro Takeuchi 1 1,Graduate School of Science and Technology, Shizuoka University,
More informationCompetitive Exclusion in a Discrete-time, Size-structured Chemostat Model
Competitive Exclusion in a Discrete-time, Size-structured Chemostat Model Hal L. Smith Department of Mathematics Arizona State University Tempe, AZ 85287 1804, USA E-mail: halsmith@asu.edu Xiao-Qiang Zhao
More informationAN ABSTRACT OF THE THESIS OF. Margaret-Rose W. Leung for the degree of Honors Baccalaureate of Science in Mathematics
AN ABSTRACT OF THE THESIS OF Margaret-Rose W. Leung for the degree of Honors Baccalaureate of Science in Mathematics presented on June 5, 2012. Title: A Vector Host Model for Coinfection by Barley/Cereal
More informationDENSITY DEPENDENCE IN DISEASE INCIDENCE AND ITS IMPACTS ON TRANSMISSION DYNAMICS
CANADIAN APPLIED MATHEMATICS QUARTERLY Volume 19, Number 3, Fall 2011 DENSITY DEPENDENCE IN DISEASE INCIDENCE AND ITS IMPACTS ON TRANSMISSION DYNAMICS REBECCA DE BOER AND MICHAEL Y. LI ABSTRACT. Incidence
More informationCoexistence of competitors in deterministic and stochastic patchy environments
0.8Copyedited by: AA 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 Journal of Biological Dynamics Vol. 00,
More informationAuthor(s) Huang, Feimin; Matsumura, Akitaka; Citation Osaka Journal of Mathematics. 41(1)
Title On the stability of contact Navier-Stokes equations with discont free b Authors Huang, Feimin; Matsumura, Akitaka; Citation Osaka Journal of Mathematics. 4 Issue 4-3 Date Text Version publisher URL
More informationOn state-space reduction in multi-strain pathogen models, with an application to antigenic drift in influenza A
On state-space reduction in multi-strain pathogen models, with an application to antigenic drift in influenza A Sergey Kryazhimskiy, Ulf Dieckmann, Simon A. Levin, Jonathan Dushoff Supporting information
More informationBifurcations in an SEIQR Model for Childhood Diseases
Bifurcations in an SEIQR Model for Childhood Diseases David J. Gerberry Purdue University, West Lafayette, IN, USA, 47907 Conference on Computational and Mathematical Population Dynamics Campinas, Brazil
More informationTowards a theory of ecotone resilience: coastal vegetation on a salinity gradient
Towards a theory of ecotone resilience: coastal vegetation on a salinity gradient Jiang Jiang, Daozhou Gao and Donald L. DeAngelis Appendix A 1.1 The model without diffusion dn 1 = N 1 ρ 1 fs 11 N 1 1
More informationA comparison of delayed SIR and SEIR epidemic models
Nonlinear Analysis: Modelling and Control, 2011, Vol. 16, No. 2, 181 190 181 A comparison of delayed SIR and SEIR epidemic models Abdelilah Kaddar a, Abdelhadi Abta b, Hamad Talibi Alaoui b a Université
More informationA simple two-patch epidemiological model with Allee effects and disease-modified fitness
Contemporary Mathematics A simple two-patch epidemiological model with Allee effects and disease-modified fitness Yun Kang and Carlos Castillo-Chavez This paper is dedicated to Professor Ronald Mickens
More informationStochastic Model for the Spread of the Hepatitis C Virus with Different Types of Virus Genome
Australian Journal of Basic and Applied Sciences, 3(): 53-65, 009 ISSN 99-878 Stochastic Model for the Spread of the Hepatitis C Virus with Different Types of Virus Genome I.A. Moneim and G.A. Mosa, Department
More informationA sharp threshold for disease persistence in host metapopulations
A sharp threshold for disease persistence in host metapopulations Thanate Dhirasakdanon, Horst R. Thieme, and P. van den Driessche Department of Mathematics and Statistics, Arizona State University, Tempe,
More informationResearch Article On the Stability Property of the Infection-Free Equilibrium of a Viral Infection Model
Hindawi Publishing Corporation Discrete Dynamics in Nature and Society Volume, Article ID 644, 9 pages doi:.55//644 Research Article On the Stability Property of the Infection-Free Equilibrium of a Viral
More informationModelling of the Hand-Foot-Mouth-Disease with the Carrier Population
Modelling of the Hand-Foot-Mouth-Disease with the Carrier Population Ruzhang Zhao, Lijun Yang Department of Mathematical Science, Tsinghua University, China. Corresponding author. Email: lyang@math.tsinghua.edu.cn,
More informationBifurcation Analysis in Simple SIS Epidemic Model Involving Immigrations with Treatment
Appl. Math. Inf. Sci. Lett. 3, No. 3, 97-10 015) 97 Applied Mathematics & Information Sciences Letters An International Journal http://dx.doi.org/10.1785/amisl/03030 Bifurcation Analysis in Simple SIS
More informationInternal Stabilizability of Some Diffusive Models
Journal of Mathematical Analysis and Applications 265, 91 12 (22) doi:1.16/jmaa.21.7694, available online at http://www.idealibrary.com on Internal Stabilizability of Some Diffusive Models Bedr Eddine
More informationCompetitive exclusion principle for SIS and SIR models with n strains
Competitive exclusion principle for SIS and SIR models with n strains Derdei Bichara, Abderrahman Iggidr, Gauthier Sallet To cite this version: Derdei Bichara, Abderrahman Iggidr, Gauthier Sallet. Competitive
More informationStability Analysis of an HIV/AIDS Epidemic Model with Screening
International Mathematical Forum, Vol. 6, 11, no. 66, 351-373 Stability Analysis of an HIV/AIDS Epidemic Model with Screening Sarah Al-Sheikh Department of Mathematics King Abdulaziz University Jeddah,
More informationMathematical Analysis of Epidemiological Models: Introduction
Mathematical Analysis of Epidemiological Models: Introduction Jan Medlock Clemson University Department of Mathematical Sciences 8 February 2010 1. Introduction. The effectiveness of improved sanitation,
More informationMathematical Epidemiology Lecture 1. Matylda Jabłońska-Sabuka
Lecture 1 Lappeenranta University of Technology Wrocław, Fall 2013 What is? Basic terminology Epidemiology is the subject that studies the spread of diseases in populations, and primarily the human populations.
More informationStability analysis of delayed SIR epidemic models with a class of nonlinear incidence rates
Published in Applied Mathematics and Computation 218 (2012 5327-5336 DOI: 10.1016/j.amc.2011.11.016 Stability analysis of delayed SIR epidemic models with a class of nonlinear incidence rates Yoichi Enatsu
More informationarxiv: v2 [q-bio.pe] 3 Oct 2018
Journal of Mathematical Biology manuscript No. (will be inserted by the editor Global stability properties of renewal epidemic models Michael T. Meehan Daniel G. Cocks Johannes Müller Emma S. McBryde arxiv:177.3489v2
More informationThreshold Conditions in SIR STD Models
Applied Mathematical Sciences, Vol. 3, 2009, no. 7, 333-349 Threshold Conditions in SIR STD Models S. Seddighi Chaharborj 1,, M. R. Abu Bakar 1, V. Alli 2 and A. H. Malik 1 1 Department of Mathematics,
More informationEvolutionary dynamics of virulence
Evolutionary dynamics of virulence Andrea Pugliese Dipartimento di Matematica Università di Trento 1 Introduction The evolution of virulence is a very active area of research; it is impossible to cite
More informationNumerical Modeling of the Transmission Dynamics of Influenza
The First International Symposium on Optimization and Systems Biology (OSB 07) Beijing, China, August 8 0, 2007 Copyright 2007 ORSC & APORC pp. 52 59 Numerical Modeling of the Transmission Dynamics of
More informationGlobal stability for a four dimensional epidemic model
Note di Matematica SSN 1123-2536, e-ssn 1590-0932 Note Mat. 30 (2010) no. 2, 83 95. doi:10.1285/i15900932v30n2p83 Global stability for a four dimensional epidemic model Bruno Buonomo Department of Mathematics
More informationMathematical Analysis of Epidemiological Models III
Intro Computing R Complex models Mathematical Analysis of Epidemiological Models III Jan Medlock Clemson University Department of Mathematical Sciences 27 July 29 Intro Computing R Complex models What
More informationChapter 3 Pullback and Forward Attractors of Nonautonomous Difference Equations
Chapter 3 Pullback and Forward Attractors of Nonautonomous Difference Equations Peter Kloeden and Thomas Lorenz Abstract In 1998 at the ICDEA Poznan the first author talked about pullback attractors of
More informationA Note on the Spread of Infectious Diseases. in a Large Susceptible Population
International Mathematical Forum, Vol. 7, 2012, no. 50, 2481-2492 A Note on the Spread of Infectious Diseases in a Large Susceptible Population B. Barnes Department of Mathematics Kwame Nkrumah University
More informationA SIMPLE SI MODEL WITH TWO AGE GROUPS AND ITS APPLICATION TO US HIV EPIDEMICS: TO TREAT OR NOT TO TREAT?
Journal of Biological Systems, Vol. 15, No. 2 (2007) 169 184 c World Scientific Publishing Company A SIMPLE SI MODEL WITH TWO AGE GROUPS AND ITS APPLICATION TO US HIV EPIDEMICS: TO TREAT OR NOT TO TREAT?
More informationSI j RS E-Epidemic Model With Multiple Groups of Infection In Computer Network. 1 Introduction. Bimal Kumar Mishra 1, Aditya Kumar Singh 2
ISSN 1749-3889 (print), 1749-3897 (online) International Journal of Nonlinear Science Vol.13(2012) No.3,pp.357-362 SI j RS E-Epidemic Model With Multiple Groups of Infection In Computer Network Bimal Kumar
More informationMathematical modelling and controlling the dynamics of infectious diseases
Mathematical modelling and controlling the dynamics of infectious diseases Musa Mammadov Centre for Informatics and Applied Optimisation Federation University Australia 25 August 2017, School of Science,
More informationA delayed SIR model with general nonlinear incidence rate
Liu Advances in Difference Equations 2015) 2015:329 DOI 10.1186/s13662-015-0619-z R E S E A R C H Open Access A delayed SIR model with general nonlinear incidence rate Luju Liu * * Correspondence: lujuliu@126.com
More informationOn the stabilizing effect of specialist predators on founder-controlled communities
On the stabilizing effect of specialist predators on founder-controlled communities Sebastian J. Schreiber Department of Mathematics Western Washington University Bellingham, WA 98225 May 2, 2003 Appeared
More informationAsynchronous and Synchronous Dispersals in Spatially Discrete Population Models
SIAM J. APPLIED DYNAMICAL SYSTEMS Vol. 7, No. 2, pp. 284 310 c 2008 Society for Industrial and Applied Mathematics Asynchronous and Synchronous Dispersals in Spatially Discrete Population Models Abdul-Aziz
More informationIMMUNE LEVEL APPROACH FOR MULTIPLE STRAIN PATHOGENS
Journal of Biological Systems, Vol. 17, No. 4 (2009) 713 737 c World Scientific Publishing Company IMMUNE LEVEL APPROACH FOR MULTIPLE STRAIN PATHOGENS M. NUÑO Department of Neurosurgery, Cedars-Sinai Medical
More informationApplied Mathematics Letters. Global behaviour of a heroin epidemic model with distributed delays
Applied Mathematics Letters 4 () 685 69 Contents lists available at ScienceDirect Applied Mathematics Letters journal homepage: www.elsevier.com/locate/aml Global behaviour of a heroin epidemic model with
More informationA Mathematical Model for the Spatial Spread of HIV in a Heterogeneous Population
A Mathematical Model for the Spatial Spread of HIV in a Heterogeneous Population Titus G. Kassem * Department of Mathematics, University of Jos, Nigeria. Abstract Emmanuel J.D. Garba Department of Mathematics,
More informationThe Existence and Stability Analysis of the Equilibria in Dengue Disease Infection Model
Journal of Physics: Conference Series PAPER OPEN ACCESS The Existence and Stability Analysis of the Equilibria in Dengue Disease Infection Model Related content - Anomalous ion conduction from toroidal
More informationModels of Infectious Disease Formal Demography Stanford Summer Short Course James Holland Jones, Instructor. August 15, 2005
Models of Infectious Disease Formal Demography Stanford Summer Short Course James Holland Jones, Instructor August 15, 2005 1 Outline 1. Compartmental Thinking 2. Simple Epidemic (a) Epidemic Curve 1:
More informationApplications in Biology
11 Applications in Biology In this chapter we make use of the techniques developed in the previous few chapters to examine some nonlinear systems that have been used as mathematical models for a variety
More informationDYNAMICS OF EVOLUTIONARY COMPETITION BETWEEN BUDDING AND LYTIC VIRAL RELEASE STRATEGIES. Xiulan Lai and Xingfu Zou. (Communicated by Yang Kuang)
MATHEMATICA IOSCIENCES doi:1.3934/mbe.214.11.191 AND ENGINEERING Volume 11, Number 5, October 214 pp. 191 1113 DYNAMICS OF EVOUTIONARY COMPETITION ETWEEN UDDING AND YTIC VIRA REEASE STRATEGIES Xiulan ai
More informationA BINOMIAL MOMENT APPROXIMATION SCHEME FOR EPIDEMIC SPREADING IN NETWORKS
U.P.B. Sci. Bull., Series A, Vol. 76, Iss. 2, 2014 ISSN 1223-7027 A BINOMIAL MOMENT APPROXIMATION SCHEME FOR EPIDEMIC SPREADING IN NETWORKS Yilun SHANG 1 Epidemiological network models study the spread
More information