The Spreading Dynamics Of Multicomponent Viruses On Complex Networks

Human history has been accompanied and shaped by various viruses such as smallpox,H1N1,H7N9,Ebola,and SARS-CoV-2.The wide spread of these viruses directly threatens people's health.Take the most recent SARS-Co V-2 virus as an example,it infected over 300 million people and caused over five million deaths around the globe in 2020 and 2021.In addition to the casualties,the in-tervention measures inevitably interrupt the functionality of the society,bringing tremendous economic burdens.Therefore,it is challenging yet crucial for re-searchers to find ways to handle the emerging viruses.The fact that viruses can transmit from one host to another makes them harder to control than other diseases.To develop effective intervention measures,it is not enough to just understand the biological mechanisms of how viruses infect hosts.Researchers also need to systematically study how the viruses are disseminated in the population,which is an important topic of network science and network epidemiology that emerged in the 21 st century.After years of research,people have accumulated rich knowledge in this regard and applied it in the fight against different outbreaks,especially in the COVID-19 pandemic.Such knowledge has helped predicting the trend of the epidemics,creating non-pharmaceutical inter-ventions,and designing vaccination strategies.However,our understanding of different viruses in mother nature is still very limited.Some viruses have very strange spreading mechanisms that differ from those that people are more familiar with,warranting further investigation.The focus,multicomponent or multipartite virus,of this dissertation is one of them.Unlike other viruses whose nucleic acid segments are encapsidated into a single viral particle that propagates as a whole,multipartite viruses have two or more segmented genomes packaged into separate virions,and each one of them can propagate independently.Still,the complete replication cycle of a multipartite virus requires the full genome so that the concurrent presence of multiple seg-ments is necessary for a successful infection.Intuitively,such mechanism hinders the spreading of multipartite viruses and puts them at an evolutionary disadvan-tage.However,studies show that about 40% of virus families have multipartite genomes.Moreover,it is shown that multipartite viruses mostly target plants as their hosts.The root of the evolutionary advantage of these viruses and their strong preference for certain types of hosts remains a mystery and draws a lot of research interests recently.This dissertation aims to characterize the spreading dynamics of multipartite viruses on complex networks through modeling and theoretical analysis and to reveal their evolutionary advantage by comparing them with normal monopartite viruses.The structure of the dissertation and the innovative contributions are listed as follows:In chapter 1,we provide a detailed introduction to the research background.In addition to a review of the network epidemiology studies,we also cover the achievements and gaps in the literature regarding multipartite viruses and lay out the research questions.This chapter is concluded with an outline of the dissertation content,research methods,and main contributions.Chapter 2 starts with an introduction of the basic knowledge closely related to this dissertation including the fundamental concepts of network science and the framework of network epidemiology.We explain the classic SIR(Susceptible-Infected-Recovered)model commonly used to describe the spreading dynamics of monopartite viruses and briefly analyze its spreading process to set the stage for the investigation below.In chapter 3,we expand the SIR model to the SLIR model(Susceptible-Latent infected-Infected-Recovered)for multipartite viruses and analyze the dynamics of SLIR model in well-mixed populations.The remaining of the dissertation is dedicated to analyzing the spreading dy-namics of multipartite viruses on networks.For the sake of simplicity,we focus on bipartite viruses.Chapter 4 considers the quench networks.Taking advantage of heterogeneous mean-field theory,we analytically solve the outbreak thresh-olds and scale of infected population.By comparing multipartite viruses with monopartite ones,we find that multipartite viruses have higher outbreak thresh-olds after holding other conditions constant.This means that it is harder for them to spread,which is consistent with the intuition.However,a key feature of mul-tipartite viruses is that their spreading processes may display discontinuous phase transitions,meaning their outbreaks are more abrupt.Chapter 5 further considers the static networks.Here we use the paired ap-proximation method to analytically solve the outbreak thresholds and scale of in-fected population.Although the outbreaks of multipartite viruses are no longer discontinuous,they are still more abrupt than those of monopartite viruses.The comparison of outbreak thresholds of different viruses yields results similar to the last chapter.In chapter 6,we compare the spreading processes of multipartite viruses on quenched and static networks.The findings suggest that multipartite viruses have smaller outbreak thresholds on static networks(that describe the structure of plant colony)than those on quenched networks when the average degree is not too small.It is the opposite for monopartite viruses.We also dive deeper into the spreading processes of multipartite viruses to reveal the mechanism behind their peculiar behaviors.Chapter 7 expands the analysis to more scenarios.First,we consider the tri-partite and quadripartite viruses.Then,we study the spreading processes of multi-partite viruses on other network structures including regular random graphs,lattice networks,scale-free networks and real-world networks.The analytical solution in these scenarios becomes intractable,we therefore rely on simulation.The main findings are consistent with the previous ones.Finally,chapter 8 summarizes the dissertation and outlines the future direc-tions.In conclusion,the present dissertation systematically analyzes the spreading dynamics of multipartite viruses on networks.To the best of our knowledge,this is the first attempt in this regard.We find that compared with their monopartite counterparts,multipartite viruses have higher outbreak thresholds,hindering their spreading.However,the outbreaks of multipartite viruses tend to be more abrupt,therefore harder to predict and control.Especially in the case of spreading on quenched networks,multipartite viruses might demonstrate discontinuous phase transitions.We further show that multipartite viruses have smaller outbreak thresh-olds on static networks when the average network degree is not too small,providing a plausible explanation of their preference for plants as their hosts.