TY - JOUR

T1 - Disease evolution on networks

T2 - the role of contact structure

AU - Read, Jonathan M.

AU - Keeling, Matt J.

PY - 2003/4/7

Y1 - 2003/4/7

N2 - Owing to their rapid reproductive rate and the severe penalties for reduced fitness, diseases are under immense evolutionary pressure. Understanding the evolutionary response of diseases in new situations has clear public-health consequences, given the changes in social and movement patterns over recent decades and the increased use of antibiotics. This paper investigates how a disease may adapt in response to the routes of transmission available between infected and susceptible individuals. The potential transmission routes are defined by a computer-generated contact network, which we describe as either local (highly clustered networks where connected individuals are likely to share common contacts) or global (unclustered networks with a high proportion of long-range connections). Evolution towards stable strategies operates through the gradual random mutation of disease traits (transmission rate and infectious period) whenever new infections occur. In contrast to mean-field models, the use of contact networks greatly constrains the evolutionary dynamics. In the local networks, high transmission rates are selected for, as there is intense competition for susceptible hosts between disease progeny. By contrast, global networks select for moderate transmission rates because direct competition between progeny is minimal and a premium is placed upon persistence. All networks show a very slow but steady rise in the infectious period.

AB - Owing to their rapid reproductive rate and the severe penalties for reduced fitness, diseases are under immense evolutionary pressure. Understanding the evolutionary response of diseases in new situations has clear public-health consequences, given the changes in social and movement patterns over recent decades and the increased use of antibiotics. This paper investigates how a disease may adapt in response to the routes of transmission available between infected and susceptible individuals. The potential transmission routes are defined by a computer-generated contact network, which we describe as either local (highly clustered networks where connected individuals are likely to share common contacts) or global (unclustered networks with a high proportion of long-range connections). Evolution towards stable strategies operates through the gradual random mutation of disease traits (transmission rate and infectious period) whenever new infections occur. In contrast to mean-field models, the use of contact networks greatly constrains the evolutionary dynamics. In the local networks, high transmission rates are selected for, as there is intense competition for susceptible hosts between disease progeny. By contrast, global networks select for moderate transmission rates because direct competition between progeny is minimal and a premium is placed upon persistence. All networks show a very slow but steady rise in the infectious period.

KW - spatial dynamics

KW - spatial heterogeneity

KW - small-world models

KW - transmission

KW - selection

KW - SMALL-WORLD NETWORKS

KW - INFECTIOUS-DISEASES

KW - COLORADO-SPRINGS

KW - SOCIAL NETWORKS

KW - VIRULENCE

KW - DYNAMICS

KW - EPIDEMICS

KW - SPREAD

KW - TRANSMISSION

KW - POPULATIONS

U2 - 10.1098/rspb.2002.2305

DO - 10.1098/rspb.2002.2305

M3 - Journal article

VL - 270

SP - 699

EP - 708

JO - Proceedings of the Royal Society B: Biological Sciences

JF - Proceedings of the Royal Society B: Biological Sciences

SN - 0962-8452

IS - 1516

ER -