A comparison of prospective space-time scan statistics and spatiotemporal event sequence based clustering for COVID-19 surveillance.
Ontology highlight
ABSTRACT:
SUBMITTER: Xu F
PROVIDER: S-EPMC8191960 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundUp to April 2022, there were six waves of infection of coronavirus disease 2019 (COVID-19) in Japan. As the outbreaks continue to grow, it is critical to detect COVID-19's clusters to allocate health resources and improve decision-making substantially. This study aimed to identify active clusters of COVID-19 in Nagasaki Prefecture and form the spatiotemporal pattern of high-risk areas in different infection periods.MethodsWe used the prospective space-time scan statistic to detect emerging COVID-19 clusters and examine the relative risk in five consecutive periods from April 1, 2020 to April 7, 2022, in Nagasaki Prefecture.ResultsThe densely inhabited districts (DIDs) in Nagasaki City have remained the most affected areas since December 2020. Most of the confirmed cases in the early period of each wave had a history of travelling to other prefectures. Community-level transmissions are suggested by the quick expansion of spatial clusters from urban areas to rural areas and remote islands. Moreover, outbreaks in welfare facilities and schools may lead to an emerging cluster in Nagasaki Prefecture's rural areas.ConclusionsThis study gives an overall analysis of the transmission dynamics of the COVID-19 pandemic in Nagasaki Prefecture, based on the number of machi-level daily cases. Furthermore, the findings in different waves can serve as references for subsequent pandemic prevention and control. This method helps the health authorities track and investigate outbreaks of COVID-19 that are specific to these environments, especially in rural areas where healthcare resources are scarce.
| S-EPMC9315091 | biostudies-literature
Project description:BACKGROUND:Overuse and misuse of prescription opioids have become signi?cant public health burdens in the United States. About 11.5 million people are estimated to have misused prescription opioids for nonmedical purposes in 2016. This has led to a signi?cant number of drug overdose deaths in the United States. Previous studies have examined spatiotemporal clusters of opioid misuse, but they have been restricted to circular shaped regions. OBJECTIVE:The goal of this study was to identify spatiotemporal hot spots of opioid users and opioid prescription claims using Medicare data. METHODS:We examined spatiotemporal clusters with signi?cantly higher number of bene?ciaries and rate of prescriptions for opioids using Medicare payment data from the Centers for Medicare & Medicaid Services. We used network scan statistics to detect signi?cant clusters with arbitrary shapes, the Kulldorff scan statistic to examine the signi?cant clusters for each year (2013, 2014, and 2015) and an expectation-based version to examine the signi?cant clusters relative to past years. Regression analysis was used to characterize the demographics of the counties that are a part of any signi?cant cluster, and data mining techniques were used to discover the specialties of the anomalous providers. RESULTS:We examined anomalous spatial clusters with respect to opioid prescription claims and bene?ciary counts and found some common patterns across states: the counties in the most anomalous clusters were fairly stable in 2014 and 2015, but they have shrunk from 2013. In Virginia, a higher percentage of African Americans in a county lower the odds of the county being anomalous in terms of opioid beneficiary counts to about 0.96 in 2015. For opioid prescription claim counts, the odds were 0.92. This pattern was consistent across the 3 states and across the 3 years. A higher number of people in the county with access to Medicaid increased the odds of the county being in the anomalous cluster to 1.16 in both types of counts in Virginia. A higher number of people with access to direct purchase of insurance plans decreased the odds of a county being in an anomalous cluster to 0.85. The expectation-based scan statistic, which captures change over time, revealed different clusters than the Kulldorff statistic. Providers with an unusually high number of opioid beneficiaries and opioid claims include specialties such as physician's assistant, nurse practitioner, and family practice. CONCLUSIONS:Our analysis of the Medicare claims data provides characteristics of the counties and provider specialties that have higher odds of being anomalous. The empirical analysis identifies highly refined spatial hot spots that are likely to encounter prescription opioid misuse and overdose. The methodology is generic and can be applied to monitor providers and their prescription behaviors in regions that are at a high risk of abuse.
| S-EPMC6601258 | biostudies-literature
Project description:BackgroundThe optimal score for ungapped local alignments of infinitely long random sequences is known to follow a Gumbel extreme value distribution. Less is known about the important case, where gaps are allowed. For this case, the distribution is only known empirically in the high-probability region, which is biologically less relevant.ResultsWe provide a method to obtain numerically the biologically relevant rare-event tail of the distribution. The method, which has been outlined in an earlier work, is based on generating the sequences with a parametrized probability distribution, which is biased with respect to the original biological one, in the framework of Metropolis Coupled Markov Chain Monte Carlo. Here, we first present the approach in detail and evaluate the convergence of the algorithm by considering a simple test case. In the earlier work, the method was just applied to one single example case. Therefore, we consider here a large set of parameters: We study the distributions for protein alignment with different substitution matrices (BLOSUM62 and PAM250) and affine gap costs with different parameter values. In the logarithmic phase (large gap costs) it was previously assumed that the Gumbel form still holds, hence the Gumbel distribution is usually used when evaluating p-values in databases. Here we show that for all cases, provided that the sequences are not too long (L > 400), a "modified" Gumbel distribution, i.e. a Gumbel distribution with an additional Gaussian factor is suitable to describe the data. We also provide a "scaling analysis" of the parameters used in the modified Gumbel distribution. Furthermore, via a comparison with BLAST parameters, we show that significance estimations change considerably when using the true distributions as presented here. Finally, we study also the distribution of the sum statistics of the k best alignments.ConclusionOur results show that the statistics of gapped and ungapped local alignments deviates significantly from Gumbel in the rare-event tail. We provide a Gaussian correction to the distribution and an analysis of its scaling behavior for several different scoring parameter sets, which are commonly used to search protein data bases. The case of sum statistics of k best alignments is included.
| S-EPMC1945026 | biostudies-literature
Project description:Coronavirus disease 2019 (COVID-19) was first identified in Wuhan, China in December 2019, and is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). COVID-19 is a pandemic with an estimated death rate between 1% and 5%; and an estimated R0 between 2.2 and 6.7 according to various sources. As of March 28th, 2020, there were over 649,000 confirmed cases and 30,249 total deaths, globally. In the United States, there were over 115,500 cases and 1891 deaths and this number is likely to increase rapidly. It is critical to detect clusters of COVID-19 to better allocate resources and improve decision-making as the outbreaks continue to grow. Using daily case data at the county level provided by Johns Hopkins University, we conducted a prospective spatial-temporal analysis with SaTScan. We detect statistically significant space-time clusters of COVID-19 at the county level in the U.S. between January 22nd-March 9th, 2020, and January 22nd-March 27th, 2020. The space-time prospective scan statistic detected "active" and emerging clusters that are present at the end of our study periods - notably, 18 more clusters were detected when adding the updated case data. These timely results can inform public health officials and decision makers about where to improve the allocation of resources, testing sites; also, where to implement stricter quarantines and travel bans. As more data becomes available, the statistic can be rerun to support timely surveillance of COVID-19, demonstrated here. Our research is the first geographic study that utilizes space-time statistics to monitor COVID-19 in the U.S.
| S-EPMC7139246 | biostudies-literature
Project description:Rapid methods for alignment-free sequence comparison make large-scale comparisons between sequences increasingly feasible. Here we study the power of the statistic D2, which counts the number of matching k-tuples between two sequences, as well as D2*, which uses centralized counts, and D2S, which is a self-standardized version, both from a theoretical viewpoint and numerically, providing an easy to use program. The power is assessed under two alternative hidden Markov models; the first one assumes that the two sequences share a common motif, whereas the second model is a pattern transfer model; the null model is that the two sequences are composed of independent and identically distributed letters and they are independent. Under the first alternative model, the means of the tuple counts in the individual sequences change, whereas under the second alternative model, the marginal means are the same as under the null model. Using the limit distributions of the count statistics under the null and the alternative models, we find that generally, asymptotically D2S has the largest power, followed by D2*, whereas the power of D2 can even be zero in some cases. In contrast, even for sequences of length 140,000 bp, in simulations D2* generally has the largest power. Under the first alternative model of a shared motif, the power of D2*approaches 100% when sufficiently many motifs are shared, and we recommend the use of D2* for such practical applications. Under the second alternative model of pattern transfer,the power for all three count statistics does not increase with sequence length when the sequence is sufficiently long, and hence none of the three statistics under consideration canbe recommended in such a situation. We illustrate the approach on 323 transcription factor binding motifs with length at most 10 from JASPAR CORE (October 12, 2009 version),verifying that D2* is generally more powerful than D2. The program to calculate the power of D2, D2* and D2S can be downloaded from http://meta.cmb.usc.edu/d2. Supplementary Material is available at www.liebertonline.com/cmb.
| S-EPMC3123933 | biostudies-literature
Project description:Large-scale comparison of the similarities between two biological sequences is a major issue in computational biology; a fast method, the D(2) statistic, relies on the comparison of the k-tuple content for both sequences. Although it has been known for some years that the D(2) statistic is not suitable for this task, as it tends to be dominated by single-sequence noise, to date no suitable adjustments have been proposed. In this article, we suggest two new variants of the D(2) word count statistic, which we call D(2)(S) and D(2)(*). For D(2)(S), which is a self-standardized statistic, we show that the statistic is asymptotically normally distributed, when sequence lengths tend to infinity, and not dominated by the noise in the individual sequences. The second statistic, D(2)(*), outperforms D(2)(S) in terms of power for detecting the relatedness between the two sequences in our examples; but although it is straightforward to simulate from the asymptotic distribution of D(2)(*), we cannot provide a closed form for power calculations.
| S-EPMC2818754 | biostudies-literature
Project description:Post-marketing drug surveillance for adverse drug events (ADEs) has typically relied on spontaneous reporting. Recently, regulatory agencies have turned their attention to more preemptive approaches that use existing data for surveillance. We conducted an environmental scan to identify active surveillance systems worldwide that use existing data for the detection of ADEs. We extracted data about the systems' structures, data, and functions. We synthesized the information across systems to identify common features of these systems. We identified nine active surveillance systems. Two systems are US based-the FDA Sentinel Initiative (including both the Mini-Sentinel Initiative and the Federal Partner Collaboration) and the Vaccine Safety Datalink (VSD); two are Canadian-the Canadian Network for Observational Drug Effect Studies (CNODES) and the Vaccine and Immunization Surveillance in Ontario (VISION); and two are European-the Exploring and Understanding Adverse Drug Reactions by Integrative Mining of Clinical Records and Biomedical Knowledge (EU-ADR) Alliance and the Vaccine Adverse Event Surveillance and Communication (VAESCO). Additionally, there is the Asian Pharmacoepidemiology Network (AsPEN) and the Shanghai Drug Monitoring and Evaluative System (SDMES). We identified two systems in the UK-the Vigilance and Risk Management of Medicines (VRMM) Division and the Drug Safety Research Unit (DSRU), an independent academic unit. These surveillance systems mostly use administrative claims or electronic medical records; most conduct pharmacovigilance on behalf of a regulatory agency. Either a common data model or a centralized model is used to access existing data. The systems have been built using national data alone or via partnership with other countries. However, active surveillance systems using existing data remain rare. North America and Europe have the most population coverage; with Asian countries making good advances.
| S-EPMC4134479 | biostudies-literature
Project description:BackgroundCOVID-19 brought enormous challenges to public health surveillance and underscored the importance of developing and maintaining robust systems for accurate surveillance. As public health data collection efforts expand, there is a critical need for infectious disease modeling researchers to continue to develop prospective surveillance metrics and statistical models to accommodate the modeling of large disease counts and variability. This paper evaluated different likelihoods for the disease count model and various spatiotemporal mean models for prospective surveillance.MethodsWe evaluated Bayesian spatiotemporal models, which are the foundation for model-based infectious disease surveillance metrics. Bayesian spatiotemporal mean models based on the Poisson and the negative binomial likelihoods were evaluated with the different lengths of past data usage. We compared their goodness of fit and short-term prediction performance with both simulated epidemic data and real data from the COVID-19 pandemic.ResultsThe simulation results show that the negative binomial likelihood-based models show better goodness of fit results than Poisson likelihood-based models as deemed by smaller deviance information criteria (DIC) values. However, Poisson models yield smaller mean square error (MSE) and mean absolute one-step prediction error (MAOSPE) results when we use a shorter length of the past data such as 7 and 3 time periods. Real COVID-19 data analysis of New Jersey and South Carolina shows similar results for the goodness of fit and short-term prediction results. Negative binomial-based mean models showed better performance when we used the past data of 52 time periods. Poisson-based mean models showed comparable goodness of fit performance and smaller MSE and MAOSPE results when we used the past data of 7 and 3 time periods.ConclusionWe evaluate these models and provide future infectious disease outbreak modeling guidelines for Bayesian spatiotemporal analysis. Our choice of the likelihood and spatiotemporal mean models was influenced by both historical data length and variability. With a longer length of past data usage and more over-dispersed data, the negative binomial likelihood shows a better model fit than the Poisson likelihood. However, as we use a shorter length of the past data for our surveillance analysis, the difference between the Poisson and the negative binomial models becomes smaller. In this case, the Poisson likelihood shows robust posterior mean estimate and short-term prediction results.
| S-EPMC10363300 | biostudies-literature
Project description:BackgroundMolecular database search tools need statistical models to assess the significance for the resulting hits. In the classical approach one asks the question how probable a certain score is observed by pure chance. Asymptotic theories for such questions are available for two random i.i.d. sequences. Some effort had been made to include effects of finite sequence lengths and to account for specific compositions of the sequences. In many applications, such as a large-scale database homology search for transmembrane proteins, these models are not the most appropriate ones. Search sensitivity and specificity benefit from position-dependent scoring schemes or use of Hidden Markov Models. Additional, one may wish to go beyond the assumption that the sequences are i.i.d. Despite their practical importance, the statistical properties of these settings have not been well investigated yet.ResultsIn this paper, we discuss an efficient and general method to compute the score distribution to any desired accuracy. The general approach may be applied to different sequence models and and various similarity measures that satisfy a few weak assumptions. We have access to the low-probability region ("tail") of the distribution where scores are larger than expected by pure chance and therefore relevant for practical applications. Our method uses recent ideas from rare-event simulations, combining Markov chain Monte Carlo simulations with importance sampling and generalized ensembles. We present results for the score statistics of fixed and random queries against random sequences. In a second step, we extend the approach to a model of transmembrane proteins, which can hardly be described as i.i.d. sequences. For this case, we compare the statistical properties of a fixed query model as well as a hidden Markov sequence model in connection with a position based scoring scheme against the classical approach.ConclusionsThe results illustrate that the sensitivity and specificity strongly depend on the underlying scoring and sequence model. A specific ROC analysis for the case of transmembrane proteins supports our observation.
| S-EPMC3042914 | biostudies-literature
Project description:Genetic correlation analysis has quickly gained popularity in the past few years and provided insights into the genetic etiology of numerous complex diseases. However, existing approaches oversimplify the shared genetic architecture between different phenotypes and cannot effectively identify precise genetic regions contributing to the genetic correlation. In this work, we introduce LOGODetect, a powerful and efficient statistical method to identify small genome segments harboring local genetic correlation signals. LOGODetect automatically identifies genetic regions showing consistent associations with multiple phenotypes through a scan statistic approach. It uses summary association statistics from genome-wide association studies (GWAS) as input and is robust to sample overlap between studies. Applied to seven phenotypically distinct but genetically correlated neuropsychiatric traits, we identify 227 non-overlapping genome regions associated with multiple traits, including multiple hub regions showing concordant effects on five or more traits. Our method addresses critical limitations in existing analytic strategies and may have wide applications in post-GWAS analysis.
| S-EPMC8016883 | biostudies-literature