Structure-Guided Identification of a Nonhuman Morbillivirus with Zoonotic Potential

Empirical and model-based evidence for a negligible role of cattle in peste des petits ruminants virus transmission and eradication

Peste des petits ruminants virus (PPRV) is a multi-host pathogen with sheep and goats as main hosts. To investigate the role of cattle in the epidemiology of PPR, we simulated conditions similar to East African zero-grazing husbandry practices in a series of trials with local Zebu cattle (Bos taurus indicus) co-housed with goats (Capra aegagrus hircus). Furthermore, we developed a mathematical model to assess the impact of PPRV-transmission from cattle to goats. Of the 32 cattle intranasally infected with the locally endemic lineage IV strain PPRV/Ethiopia/Habru/2014 none transmitted PPRV to 32 co-housed goats. However, these cattle or cattle co-housed with PPRV-infected goats seroconverted. The results confirm previous studies that cattle currently play a negligible role in PPRV-transmission and small ruminant vaccination is sufficient for eradication. However, the possible emergence of PPRV strains more virulent for cattle may impact eradication. Therefore, continued monitoring of PPRV circulation and evolution is recommended.

Comparative evolutionary analyses of peste des petits ruminants virus genetic lineages

Peste des petits ruminants virus (PPRV) causes a highly infectious disease affecting mainly goats and sheep in large parts of Africa, Asia, and the Middle East and has an important impact on the global economy and food security. Full genome sequencing of PPRV strains has proved to be critical to increasing our understanding of PPR epidemiology and to inform the ongoing global efforts for its eradication. However, the number of full PPRV genomes published is still limited and with a heavy bias towards recent samples and genetic Lineage IV (LIV), which is only one of the four existing PPRV lineages. Here, we generated genome sequences for twenty-five recent (2010-6) and seven historical (1972-99) PPRV samples, focusing mainly on Lineage II (LII) in West Africa. This provided the first opportunity to compare the evolutionary pressures and history between the globally dominant PPRV genetic LIV and LII, which is endemic in West Africa. Phylogenomic analysis showed that the relationship between PPRV LII strains was complex and supported the extensive transboundary circulation of the virus within West Africa. In contrast, LIV sequences were clearly separated per region, with strains from West and Central Africa branched as a sister clade to all other LIV sequences, suggesting that this lineage also has an African origin. Estimates of the time to the most recent common ancestor place the divergence of modern LII and LIV strains in the 1960s-80s, suggesting that this period was particularly important for the diversification and spread of PPRV globally. Phylogenetic relationships among historical samples from LI, LII, and LIII and with more recent samples point towards a high genetic diversity for all these lineages in Africa until the 1970s-80s and possible bottleneck events shaping PPRV's evolution during this period. Molecular evolution analyses show that strains belonging to LII and LIV have evolved under different selection pressures. Differences in codon usage and adaptative selection pressures were observed in all viral genes between the two lineages. Our results confirm that comparative genomic analyses can provide new insights into PPRV's evolutionary history and molecular epidemiology. However, PPRV genome sequencing efforts must be ramped up to increase the resolution of such studies for their use in the development of efficient PPR control and surveillance strategies.

Pentagon Found Daily, Metagenomic Detection of Novel Bioaerosol Threats to Be Cost-Prohibitive: Can Virtualization and AI Make It Cost-Effective?

In 2022, the Pentagon Force Protection Agency found threat agnostic detection of novel bioaerosol threats to be "not feasible for daily operations" due to the cost of reagents used for metagenomics, cost of sequencing instruments, and cost of labor for subject matter experts to analyze bioinformatics. Similar operational difficulties might extend to many of the 280,000 buildings (totaling 2.3 billion square feet) at 5,000 secure US Department of Defense military sites, 250 Navy ships, as well as many civilian buildings. These economic barriers can still be addressed in a threat agnostic manner by dynamically pooling samples from dry filter units, called spike-triggered virtualization, whereby pooling and sequencing depth are automatically modulated based on novel biothreats in the sequencing output. By running at a high average pooling factor, the daily and annual cost per dry filter unit can be reduced by 10 to 100 times depending on the chosen trigger thresholds. Artificial intelligence can further enhance the sensitivity of spike-triggered virtualization. The risk of infection during the 12- to 24-hour window between a bioaerosol incident and its detection remains, but in some cases it can be reduced by 80% or more with high-speed indoor air cleaning exceeding 12 air changes per hour, which is similar to the rate of air cleaning in passenger airplanes in flight. That level of air changes per hour or higher is likely to be cost-prohibitive using central heating ventilation and air conditioning systems, but it can be achieved economically by using portable air filtration in rooms with typical ceiling heights (less than 10 feet) for a cost of approximately $0.50 to $1 per square foot for do-it-yourself units and $2 to $5 per square foot for high-efficiency particulate air filters.

The (Re-)Emergence and Spread of Viral Zoonotic Disease: A Perfect Storm of Human Ingenuity and Stupidity

Diseases that are transmitted from vertebrate animals to humans are referred to as zoonotic diseases. Although microbial agents such as bacteria and parasites are linked to zoonotic events, viruses account for a high percentage of zoonotic diseases that have emerged. Worryingly, the 21st century has seen a drastic increase in the emergence and re-emergence of viral zoonotic disease. Even though humans and animals have coexisted for millennia, anthropogenic factors have severely increased interactions between the two populations, thereby increasing the risk of disease spill-over. While drivers such as climate shifts, land exploitation and wildlife trade can directly affect the (re-)emergence of viral zoonotic disease, globalisation, geopolitics and social perceptions can directly facilitate the spread of these (re-)emerging diseases. This opinion paper discusses the "intelligent" nature of viruses and their exploitation of the anthropogenic factors driving the (re-)emergence and spread of viral zoonotic disease in a modernised and connected world.

Morbillivirus: A highly adaptable viral genus

Over the course of human history, numerous diseases have been caused by the transmission of viruses from an animal reservoir into the human population. The viruses of the genus Morbillivirus are human and animal pathogens that emerged from a primordial ancestor a millennia ago and have been transmitting to new hosts, adapting, and evolving ever since. Through interaction with susceptible individuals, as yet undiscovered morbilliviruses or existing morbilliviruses in animal hosts could cause future zoonotic diseases in humans.

Paramyxoviruses from bats: changes in receptor specificity and their role in host adaptation

Global metagenomic surveys have revealed that bats host a diverse array of paramyxoviruses, including species from at least five major genera. An essential determinant of successful spillover is the entry of a virus into a new host. We evaluate the role of receptor usage in the zoonotic potential of bat-borne henipaviruses, morbilliviruses, pararubulaviruses, orthorubulaviruses, and jeilongviruses; successful spillover into humans depends upon compatibility of a respective viral attachment protein with its cognate receptor. We also emphasize the importance of postentry restrictions in preventing spillover. Metagenomics and characterization of newly identified paramyxoviruses have greatly improved our understanding of spillover determinants, allowing for better forecasts of which bat-borne viruses may pose the greatest risk for cross-species transmission into humans.

Novel and classical morbilliviruses: Current knowledge of three divergent morbillivirus groups

Currently, seven species of morbillivirus have been classified. Six of these species (Measles morbillivirus, Rinderpest morbillivirus, Small ruminant morbillivirus, Canine morbillivirus, Phocine morbillivirus, and Cetacean morbillivirus) are highly infectious and cause serious systemic diseases in humans, livestock, domestic dogs, and wild animals. These species commonly use the host proteins signaling lymphocytic activation molecule (SLAM) and nectin-4 as receptors, and this usage contributes to their virulence. The seventh species (Feline morbillivirus: FeMV) is phylogenetically divergent from the six SLAM-using species. FeMV differs from the SLAM-using morbillivirus group in pathogenicity and infectivity, and is speculated to use non-SLAM receptors. Recently, novel species of morbilliviruses have been discovered in bats, rodents, and domestic pigs. Because the ability to use SLAM and nectin-4 is closely related to the infectivity and pathogenicity of morbilliviruses, investigation of the potential usage of these receptors is useful for estimating infectivity and pathogenicity. The SLAM-binding sites in the receptor-binding protein show high similarity among the SLAM-using morbilliviruses. This feature may help to estimate whether novel morbillivirus species can use SLAM as a receptor. A novel morbillivirus species isolated from wild mice diverged from the classified morbilliviruses in the phylogenetic tree, forming a third group separate from the SLAM-using morbillivirus group and FeMV. This suggests that the novel rodent morbillivirus may exhibit a different risk from the SLAM-using morbillivirus group, and analyses of its viral pathogenicity and infectivity toward humans are warranted.

Challenges in modeling the emergence of novel pathogens

The emergence of infectious agents with pandemic potential present scientific challenges from detection to data interpretation to understanding determinants of risk and forecasts. Mathematical models could play an essential role in how we prepare for future emergent pathogens. Here, we describe core directions for expansion of the existing tools and knowledge base, including: using mathematical models to identify critical directions and paths for strengthening data collection to detect and respond to outbreaks of novel pathogens; expanding basic theory to identify infectious agents and contexts that present the greatest risks, over both the short and longer term; by strengthening estimation tools that make the most use of the likely range and uncertainties in existing data; and by ensuring modelling applications are carefully communicated and developed within diverse and equitable collaborations for increased public health benefit.

Application of error-prone PCR to functionally probe the morbillivirus Haemagglutinin protein

The enveloped morbilliviruses utilise conserved proteinaceous receptors to enter host cells: SLAMF1 or Nectin-4. Receptor binding is initiated by the viral attachment protein Haemagglutinin (H), with the viral Fusion protein (F) driving membrane fusion. Crystal structures of the prototypic morbillivirus measles virus H with either SLAMF1 or Nectin-4 are available and have served as the basis for improved understanding of this interaction. However, whether these interactions remain conserved throughout the morbillivirus genus requires further characterisation. Using a random mutagenesis approach, based on error-prone PCR, we targeted the putative receptor binding site for SLAMF1 interaction on peste des petits ruminants virus (PPRV) H, identifying mutations that inhibited virus-induced cell-cell fusion. These data, combined with structural modelling of the PPRV H and ovine SLAMF1 interaction, indicate this region is functionally conserved across all morbilliviruses. Error-prone PCR provides a powerful tool for functionally characterising functional domains within viral proteins.

Mutation Profiles of eGFP-Tagged Small Ruminant Morbillivirus During 45 Serial Passages in Ribavirin-Treated Cells

Small ruminant morbillivirus (SRMV), formerly known as peste-des-petits-ruminants virus, classified into the genus Morbillivirus in the family Paramyxoviridae. Its L protein functions as the RNA-dependent RNA polymerases (RdRp) during viral replication. Due to the absence of efficient proofreading activity in their RdRps, various RNA viruses reveal high mutation frequencies, making them evolve rapidly during serial passages in cells, especially treated with a certain mutagen, like ribavirin. We have previously rescued a recombinant enhanced green fluorescence protein-tagged SRMV (rSRMV-eGFP) using reverse genetics. In this study, the rSRMV-eGFP was subjected to serial passages in ribavirin-treated cells. Due to the ribavirin-exerted selective pressure, it was speculated that viral progenies would form quasispecies after dozens of passages. Viral progenies at passage-10, -20, -30, -40, and -50 were separately subjected to next-generation sequencing (NGS), consequently revealing a total of 34 single-nucleotide variations, including five synonymous, 21 missense, and one non-sense mutations. The L sequence was found to harbor eight missense mutations during serial passaging. It was speculated that at least one high-fidelity variant was present in viral quasispecies at passage-50. If purified from the population of viral progenies, this putative variant would contribute to clarifying a molecular mechanism in viral high-fidelity replication in vitro.

Molecular epidemiology of peste des petits ruminants virus emergence in critically endangered Mongolian saiga antelope and other wild ungulates

Peste des petits ruminants virus (PPRV) causes disease in domestic and wild ungulates, is the target of a Global Eradication Programme, and threatens biodiversity. Understanding the epidemiology and evolution of PPRV in wildlife is important but hampered by the paucity of wildlife-origin PPRV genomes. In this study, full PPRV genomes were generated from three Mongolian saiga antelope, one Siberian ibex, and one goitered gazelle from the 2016-2017 PPRV outbreak. Phylogenetic analysis showed that for Mongolian and Chinese PPRV since 2013, the wildlife and livestock-origin genomes were closely related and interspersed. There was strong phylogenetic support for a monophyletic group of PPRV from Mongolian wildlife and livestock, belonging to a clade of lineage IV PPRV from livestock and wildlife from China since 2013. Discrete diffusion analysis found strong support for PPRV spread into Mongolia from China, and phylogeographic analysis indicated Xinjiang Province as the most likely origin, although genomic surveillance for PPRV is poor and lack of sampling from other regions could bias this result. Times of most recent common ancestor (TMRCA) were June 2015 (95 per cent highest posterior density (HPD): August 2014 to March 2016) for all Mongolian PPRV genomes and May 2016 (95 per cent HPD: October 2015 to October 2016) for Mongolian wildlife-origin PPRV. This suggests that PPRV was circulating undetected in Mongolia for at least 6 months before the first reported outbreak in August 2016 and that wildlife were likely infected before livestock vaccination began in October 2016. Finally, genetic variation and positively selected sites were identified that might be related to PPRV emergence in Mongolian wildlife. This study is the first to sequence multiple PPRV genomes from a wildlife outbreak, across several host species. Additional full PPRV genomes and associated metadata from the livestock-wildlife interface are needed to enhance the power of molecular epidemiology, support PPRV eradication, and safeguard the health of the whole ungulate community.

Mink SLAM V-Region V74I Substitutions Contribute to the Formation of Syncytia Induced by Canine Distemper Virus

The Signal lymphatic activation molecule (SLAM, also known as CD150) as the cellular receptor of canine distemper virus (CDV) plays an important role in the virus-host interaction. However, it is still unknown whether amino acid differences in the SLAM variable (V) region affect the formation of syncytia. Here, using raccoon dog SLAM (rSLAM) and mink SLAM (mSLAM), we performed SLAM-V homologous modeling, site-directed mutagenesis, and surface expression analysis, as well as a cell fusion assay, to study the interaction between SLAM and CDV. More specifically, our investigation focused on two amino acid residues (74 and 129) of SLAM, previously predicted to play a relevant role in receptor-ligand interaction. Our results indicated that only residues at position 60, 74, and 129 were different between rSLAM and mSLAM among the 29 amino acids that might interact with CDV H, and residues 74 and 129 were located in the interface region interacting with CDV H. The amino acid substitution at the positions of 74 have a significant effect on the expression of mSLAM. The SLAM-V74I mutation in mink significantly improved the cell fusion efficiency of CDV. In contrast, the SLAM-I74V mutation in the raccoon dog significantly decreased cell fusion efficiency. We conclude that residue 74 of SLAM plays an important role during the the formation of syncytia. Only when implementing CDV infection analysis, the rSLAM-Q129R can significantly decreased the mean number of syncytia, but the mSLAM-R129Q can't. Additionally, residue 60 show variability between rSLAM and mSLAM. We believe that our study makes a significant contribution to the literature because we provide molecular data, partially accounting for the differences in host membrane and virus interaction laying the foundation for further molecular work.

Micro-fusion inhibition tests: quantifying antibody neutralization of virus-mediated cell-cell fusion

Although enveloped viruses canonically mediate particle entry through virus-cell fusion, certain viruses can spread by cell-cell fusion, brought about by receptor engagement and triggering of membrane-bound, viral-encoded fusion proteins on the surface of cells. The formation of pathogenic syncytia or multinucleated cells is seen in vivo, but their contribution to viral pathogenesis is poorly understood. For the negative-strand paramyxoviruses respiratory syncytial virus (RSV) and Nipah virus (NiV), cell-cell spread is highly efficient because their oligomeric fusion protein complexes are active at neutral pH. The recently emerged severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has also been reported to induce syncytia formation in infected cells, with the spike protein initiating cell-cell fusion. Whilst it is well established that fusion protein-specific antibodies can block particle attachment and/or entry into the cell (canonical virus neutralization), their capacity to inhibit cell-cell fusion and the consequences of this neutralization for the control of infection are not well characterized, in part because of the lack of specific tools to assay and quantify this activity. Using an adapted bimolecular fluorescence complementation assay, based on a split GFP-Renilla luciferase reporter, we have established a micro-fusion inhibition test (mFIT) that allows the identification and quantification of these neutralizing antibodies. This assay has been optimized for high-throughput use and its applicability has been demonstrated by screening monoclonal antibody (mAb)-mediated inhibition of RSV and NiV fusion and, separately, the development of fusion-inhibitory antibodies following NiV vaccine immunization in pigs. In light of the recent emergence of coronavirus disease 2019 (COVID-19), a similar assay was developed for SARS-CoV-2 and used to screen mAbs and convalescent patient plasma for fusion-inhibitory antibodies. Using mFITs to assess antibody responses following natural infection or vaccination is favourable, as this assay can be performed entirely at low biocontainment, without the need for live virus. In addition, the repertoire of antibodies that inhibit cell-cell fusion may be different to those that inhibit particle entry, shedding light on the mechanisms underpinning antibody-mediated neutralization of viral spread.

The SARS-CoV-2 Spike protein has a broad tropism for mammalian ACE2 proteins

SARS Coronavirus 2 (SARS-CoV-2) emerged in late 2019, leading to the Coronavirus Disease 2019 (COVID-19) pandemic that continues to cause significant global mortality in human populations. Given its sequence similarity to SARS-CoV, as well as related coronaviruses circulating in bats, SARS-CoV-2 is thought to have originated in Chiroptera species in China. However, whether the virus spread directly to humans or through an intermediate host is currently unclear, as is the potential for this virus to infect companion animals, livestock, and wildlife that could act as viral reservoirs. Using a combination of surrogate entry assays and live virus, we demonstrate that, in addition to human angiotensin-converting enzyme 2 (ACE2), the Spike glycoprotein of SARS-CoV-2 has a broad host tropism for mammalian ACE2 receptors, despite divergence in the amino acids at the Spike receptor binding site on these proteins. Of the 22 different hosts we investigated, ACE2 proteins from dog, cat, and cattle were the most permissive to SARS-CoV-2, while bat and bird ACE2 proteins were the least efficiently used receptors. The absence of a significant tropism for any of the 3 genetically distinct bat ACE2 proteins we examined indicates that SARS-CoV-2 receptor usage likely shifted during zoonotic transmission from bats into people, possibly in an intermediate reservoir. Comparison of SARS-CoV-2 receptor usage to the related coronaviruses SARS-CoV and RaTG13 identified distinct tropisms, with the 2 human viruses being more closely aligned. Finally, using bioinformatics, structural data, and targeted mutagenesis, we identified amino acid residues within the Spike-ACE2 interface, which may have played a pivotal role in the emergence of SARS-CoV-2 in humans. The apparently broad tropism of SARS-CoV-2 at the point of viral entry confirms the potential risk of infection to a wide range of companion animals, livestock, and wildlife.

Will an outbreak exceed available resources for control? Estimating the risk from invading pathogens using practical definitions of a severe epidemic

Forecasting whether or not initial reports of disease will be followed by a severe epidemic is an important component of disease management. Standard epidemic risk estimates involve assuming that infections occur according to a branching process and correspond to the probability that the outbreak persists beyond the initial stochastic phase. However, an alternative assessment is to predict whether or not initial cases will lead to a severe epidemic in which available control resources are exceeded. We show how this risk can be estimated by considering three practically relevant potential definitions of a severe epidemic; namely, an outbreak in which: (i) a large number of hosts are infected simultaneously; (ii) a large total number of infections occur; and (iii) the pathogen remains in the population for a long period. We show that the probability of a severe epidemic under these definitions often coincides with the standard branching process estimate for the major epidemic probability. However, these practically relevant risk assessments can also be different from the major epidemic probability, as well as from each other. This holds in different epidemiological systems, highlighting that careful consideration of how to classify a severe epidemic is vital for accurate epidemic risk quantification.

Camelids and Cattle Are Dead-End Hosts for Peste-des-Petits-Ruminants Virus

Peste-des-petits-ruminants virus (PPRV) causes a severe respiratory disease in small ruminants. The possible impact of different atypical host species in the spread and planed worldwide eradication of PPRV remains to be clarified. Recent transmission trials with the virulent PPRV lineage IV (LIV)-strain Kurdistan/2011 revealed that pigs and wild boar are possible sources of PPRV-infection. We therefore investigated the role of cattle, llamas, alpacas, and dromedary camels in transmission trials using the Kurdistan/2011 strain for intranasal infection and integrated a literature review for a proper evaluation of their host traits and role in PPRV-transmission. Cattle and camelids developed no clinical signs, no viremia, shed no or only low PPRV-RNA loads in swab samples and did not transmit any PPRV to the contact animals. The distribution of PPRV-RNA or antigen in lymphoid organs was similar in cattle and camelids although generally lower compared to suids and small ruminants. In the typical small ruminant hosts, the tissue tropism, pathogenesis and disease expression after PPRV-infection is associated with infection of immune and epithelial cells via SLAM and nectin-4 receptors, respectively. We therefore suggest a different pathogenesis in cattle and camelids and both as dead-end hosts for PPRV.

Basal interferon signaling and therapeutic use of interferons in controlling peste des petits ruminants virus infection

Peste des petits ruminants virus (PPRV) is a morbillivirus which causes severe disease in ruminants. Since interferons (IFNs) serve as the important defense line against viral infection, we have investigated the roles of types I and III IFNs in PPRV infection in vitro. Upon PPRV infection, IFN-λ3 was strongly induced, while IFN-β and IFN-λ2 were moderately induced at transcriptional level in human embryonic kidney 293 T (HEK293T) cells. Although the transcription of type I and III IFNs were triggered, the production of functional IFN products was not detected. Importantly, the replication of PPRV was strongly inhibited in HEK293T cells treated by the exogenous IFNs (IFN-α-2b, IFN-β and IFN-λ3). Consistently, these IFNs significantly activate a panel of IFN-stimulated genes (ISGs). The inhibition of JAK-STAT pathway by JAK I inhibitor can abrogate the anti-PPRV activity of IFNs. Thus, our study shall contribute to better understanding of the complex PPRV-host interactions and provide rationale for therapeutic development of IFN-based treatment against PPRV infection.

Probing Morbillivirus Antisera Neutralization Using Functional Chimerism between Measles Virus and Canine Distemper Virus Envelope Glycoproteins

Measles virus (MeV) is monotypic. Live virus challenge provokes a broadly protective humoral immune response that neutralizes all known measles genotypes. The two surface glycoproteins, H and F, mediate virus attachment and entry, respectively, and neutralizing antibodies to H are considered the main correlate of protection. Herein, we made improvements to the MeV reverse genetics system and generated a panel of recombinant MeVs in which the globular head domain or stalk region of the H glycoprotein or the entire F protein, or both, were substituted with the corresponding protein domains from canine distemper virus (CDV), a closely related morbillivirus that resists neutralization by measles-immune sera. The viruses were tested for sensitivity to human or guinea pig neutralizing anti-MeV antisera and to ferret anti-CDV antisera. Virus neutralization was mediated by antibodies to both H and F proteins, with H being immunodominant in the case of MeV and F being so in the case of CDV. Additionally, the globular head domains of both MeV and CDV H proteins were immunodominant over their stalk regions. These data shed further light on the factors constraining the evolution of new morbillivirus serotypes.

Marine Morbilliviruses: Diversity and Interaction with Signaling Lymphocyte Activation Molecules

Epidemiological reports of phocine distemper virus (PDV) and cetacean morbillivirus (CeMV) have accumulated since their discovery nearly 30 years ago. In this review, we focus on the interaction between these marine morbilliviruses and their major cellular receptor, the signaling lymphocyte activation molecule (SLAM). The three-dimensional crystal structure and homology models of SLAMs have demonstrated that 35 residues are important for binding to the morbillivirus hemagglutinin (H) protein and contribute to viral tropism. These 35 residues are essentially conserved among pinnipeds and highly conserved among the Caniformia, suggesting that PDV can infect these animals, but are less conserved among cetaceans. Because CeMV can infect various cetacean species, including toothed and baleen whales, the CeMV-H protein is postulated to have broader specificity to accommodate more divergent SLAM interfaces and may enable the virus to infect seals. In silico analysis of viral H protein and SLAM indicates that each residue of the H protein interacts with multiple residues of SLAM and vice versa. The integration of epidemiological, virological, structural, and computational studies should provide deeper insight into host specificity and switching of marine morbilliviruses.

Period of Measurement in Time-Series Predictions of Disease Counts from 2007 to 2017 in Northern Nevada: Analytics Experiment

BACKGROUND

The literature in statistics presents methods by which autocorrelation can identify the best period of measurement to improve the performance of a time-series prediction. The period of measurement plays an important role in improving the performance of disease-count predictions. However, from the operational perspective in public health surveillance, there is a limitation to the length of the measurement period that can offer meaningful and valuable predictions.

OBJECTIVE

This study aimed to establish a method that identifies the shortest period of measurement without significantly decreasing the prediction performance for time-series analysis of disease counts.

METHODS

The data used in this evaluation include disease counts from 2007 to 2017 in northern Nevada. The disease counts for chlamydia, salmonella, respiratory syncytial virus, gonorrhea, viral meningitis, and influenza A were predicted.

RESULTS

Our results showed that autocorrelation could not guarantee the best performance for prediction of disease counts. However, the proposed method with the change-point analysis suggests a period of measurement that is operationally acceptable and performance that is not significantly different from the best prediction.

CONCLUSIONS

The use of change-point analysis with autocorrelation provides the best and most practical period of measurement.