Transmission dynamics and molecular epidemiology of arboviruses in Indonesia

Large scale outbreaks of dengue virus (DENV) and dengue haemorrhagic fever are reported every year in Indonesia. In addition, Chikungunya virus (CHIKV) is endemic in parts of Indonesia, and two positive cases of Zika virus (ZIKV) have been described, one from an Australian who travelled to Indonesia, a second from Jambi province. DENV, CHIKV, and ZIKV are all transmitted by the mosquito Aedes aegypti, and are co-circulating in the same population. They also cause overlapping symptoms in adults, and hence are difficult to differentiate by symptoms alone. Indonesia has a large population - around 250 million people - with around 90% of the population living in areas suitable for the transmission of 'arboviruses' (short for arthropod-borne viruses), and collectively, arboviruses represent a major public health concern in Indonesia. The number of cases varies both geographically and over time, and relatively little is known about the drivers of this variation in Indonesia.

DENV, CHIKV, and ZIKV are all RNA viruses, which evolve at a high rate. Consequently, these viruses are genetically variable, and through the application of statistical models to viral sequence data, information on the dynamics of transmission, both over time and space, can be obtained. However, although the number of arbovirus infections per year is high, the number of viral sequences from Indonesia is relatively low, and often, these sequences only span a small part of the viral genome.

This project aims to characterise the transmission dynamics of DENV, CHIKV, and ZIKV through viral sequencing of clinical samples from humans. We will harness a network of collaborators to provide archived samples, and prospective studies in locations representing west (Sumatra - Palembang City), central (Kalimantan - Banjarmasin city), and east (Maluku - Ambon city) Indonesia. In addition, we will study outbreaks that may occur during the study period. Using next-generation sequencing platforms, we aim to generate a large dataset of full-length viral genomes that will significantly add to the existing body of viral sequence data currently available.

Sequences of DENV and CHIKV (and potentially ZIKV) will be analysed using a 'phylodynamic' approach, integrating the sequence data with epidemiological data using mathematical models in order to gain a more complete understanding of the transmission dynamics of these arboviruses.

Arthropod-borne viruses, or arboviruses, are becoming an increasing public health problem, especially in the tropics and sub-tropics. While Indonesia suffers a high burden of dengue and Chikungunya, relatively little is known about the epidemiology of these infections. As dengue virus (DENV) and Chikungunya virus (CHIKV) are rapidly-evolving RNA viruses, viral sequence data can offer insights into transmission, even with relatively limited samples. However, to date arboviral sequence data is limited; the number of sequences is relatively low, and sequences are often short, spanning only part of the viral genome. As such, the existing body of sequence data provides relatively little information on the transmission dynamics of arboviruses.

The aim of this project is to use sequence data of DENV and CHIKV in order to better understand the dynamics of these co-circulating pathogens, over both space and time. We will compare and contrast the use of two next-generation sequencing platforms, the Ion Torrent Personal Genome Machine and the MinION nanopore sequencer, to generate near-full length genomes of DENV and CHIKV, as well as ZIKV, if positive samples are detected. Archived samples are available to allow a comparison of these platforms with conventional Sanger sequencing. We will pilot the use of these platforms in a prospective, hospital-based study in three areas of Indonesia. Individuals presenting at these hospitals with symptoms of dengue fever will be tested for DENV, CHIKV, and ZIKV, and viral sequences generated from positive samples. In order to link the viral sequence data with classical epidemiological data, we will build 'phylodynamic' models, which consider the relationship between transmission events and the viral phylogeny. This proposal will not only provide information on the molecular epidemiology of DENV and CHIKV, but also increase the capacity in bioinformatics and mathematical modeling in Indonesia.

Academics:
a) Other researchers in the field of arbovirus infection, emerging viruses, RNA viruses, viral evolution and phylodynamics will benefit. The data generated in the project may be used for comparison studies of the same viruses from different regions and other RNA viruses. Epidemiology and transmission models will be applicable to other viruses that cause a public health concern. This will be realized after deposition of data in databases, software to GITHub and publication of results, which will occur in the short to medium term (6 months to 2 years).
b) Indonesian scientists will benefit from training in modern sequencing and analysis techniques that they can then use independently to continue to monitor arbovirus infection prevalence and spread in Indonesia. These techniques can be applied to many different infectious diseases and so will allow the scientists to investigate other diseases of importance in Indonesia. This will be available immediately on training of individuals.
c) The post-doctoral research scientist employed through the project will gain skills as above for the Indonesian scientists but will also be able to take part in the University of Cambridge transferable skills training available to all University employees. This includes presentation, outreach training and entrepreneurial training.

Public Health Professionals:
a) Indonesian public health professionals will benefit from the information on viral prevalence, coinfection and ability of different strains of virus to transmit. This will allow them to formulate public health measures specific for Indonesia. These measures would be applicable within the short term ie within 3 months should major changes in virus prevalence and transmissibility be detected.
b) World wide public health bodies (including PHE and CDC) will benefit from viral prevalence data allowing travel health advice to be given. Again, this would be actioned more rapidly should changes in virus prevalence and transmissibility be detected.

Public:
The general Indonesian population and tourists/travelers will benefit from viral prevalence data and more specific diagnosis of infection. This will allow better avoidance of infection and treatment should infection occur. This should decrease loss of working days and better responses to specific treatments. With prospective surveillance, virology data will be available within a week of samples being taken for clinical cases of illness.

Commercial:
Vaccine and drug companies will have access to the sequence data to allow development of novel vaccines and possible antivirals. This is a long term benefit and will require buy-in by companies to develop the reagents. As the data and viral infections are associated with developing countries, there needs to be a commercial case made or academic/commercial partnership for this to develop. The data will be particularly relevant for virus strains local to Indonesia.