Systems-based approach for optimization of a scalable bacterial ST mapping assembly-free algorithm

Epidemiological surveillance of bacterial pathogens requires real-time data analysis with a fast turn-around, while aiming at generating two main outcomes: 1) Species level identification; and 2) Variant mapping at different levels of genotypic resolution for population-based tracking, in addition to predicting traits such as antimicrobial resistance (AMR). With the recent advances and continual dissemination of whole-genome sequencing technologies, large-scale population-based genotyping of bacterial pathogens has become possible. Since bacterial populations often present a high degree of clonality in the genomic backbone (i.e., low genetic diversity), the choice of genotyping scheme can even facilitate the understanding of ancestral relationships and can be used for prediction of co-inherited traits such as AMR. Multi-locus sequence typing (MLST) fits that purpose and can identify sequence types (ST) based on seven ubiquitous genome-scattered loci that aid in genotyping isolates beneath the species level. ST-based mapping also standardizes genotyping across laboratories and can be consistently used worldwide. However, ST-based algorithms, when using Illumina paired-end sequences, often rely on genome assembly prior to classification. That hinders rapid genotyping and scalability which are essential aspects of genomic epidemiology. stringMLST is a kmer-based ST method with the capacity to solve both hurdles. Yet, a comprehensive scalable comparison of its use in contrast to a standard MLST program for a wide array of phylogenetically divergent Public Health-relevant bacterial pathogens is lacking. Herein, we first demonstrated that stringMLST is a fast tool that can be deployed for ST-based epidemiological inquiries of bacterial populations. Additionally, we systematically evaluated and showed the impact of genome-intrinsic and -extrinsic features, as well as the optimal kmer length in maximizing the performance of stringMLST on species-by-species basis, and highlighted a few instances where this program may not be applicable in its current format. Furthermore, we integrated stringMLST as part of our freely available and scalable hierarchical-based population genomics platform called ProkEvo. Besides facilitating automatable and reproducible bacterial population guided analysis, ProkEvo now offers a rapidly deployable genomic epidemiology tool for ST mapping, with specific guidance on how to optimize its performance, that can be widely applicable by microbiological laboratories and epidemiological agencies.

Download Full-text

Multi-GPU approach to global induction of classification trees for large-scale data mining

Applied Intelligence ◽

10.1007/s10489-020-01952-5 ◽

2021 ◽

Author(s):

Krzysztof Jurczuk ◽

Marcin Czajkowski ◽

Marek Kretowski

Keyword(s):

Data Mining ◽

Large Scale ◽

Real Life ◽

Population Based ◽

Tree Structure ◽

Global Approach ◽

Data Parallel ◽

Large Scale Data ◽

The Impact ◽

Scale Data

AbstractThis paper concerns the evolutionary induction of decision trees (DT) for large-scale data. Such a global approach is one of the alternatives to the top-down inducers. It searches for the tree structure and tests simultaneously and thus gives improvements in the prediction and size of resulting classifiers in many situations. However, it is the population-based and iterative approach that can be too computationally demanding to apply for big data mining directly. The paper demonstrates that this barrier can be overcome by smart distributed/parallel processing. Moreover, we ask the question whether the global approach can truly compete with the greedy systems for large-scale data. For this purpose, we propose a novel multi-GPU approach. It incorporates the knowledge of global DT induction and evolutionary algorithm parallelization together with efficient utilization of memory and computing GPU’s resources. The searches for the tree structure and tests are performed simultaneously on a CPU, while the fitness calculations are delegated to GPUs. Data-parallel decomposition strategy and CUDA framework are applied. Experimental validation is performed on both artificial and real-life datasets. In both cases, the obtained acceleration is very satisfactory. The solution is able to process even billions of instances in a few hours on a single workstation equipped with 4 GPUs. The impact of data characteristics (size and dimension) on convergence and speedup of the evolutionary search is also shown. When the number of GPUs grows, nearly linear scalability is observed what suggests that data size boundaries for evolutionary DT mining are fading.

Download Full-text

The impact of C-reactive protein levels on headache frequency in the HUNT study 2006–2008

BMC Neurology ◽

10.1186/s12883-019-1462-8 ◽

2019 ◽

Vol 19 (1) ◽

Cited By ~ 3

Author(s):

Knut Hagen ◽

Lars Jacob Stovner ◽

Kristian Bernhard Nilsen ◽

Espen Saxhaug Kristoffersen ◽

Bendik Slagsvold Winsvold

Keyword(s):

Migraine With Aura ◽

Large Scale ◽

Population Based ◽

C Reactive Protein ◽

Cross Sectional ◽

Population Based Study ◽

Reactive Protein ◽

Scale Population ◽

Hs Crp ◽

The Impact

Abstract Background Increased high sensitivity C- reactive protein (hs-CRP) levels have been found in many earlier studies on migraine, and recently also in persons with migraine and insomnia. The aim of this study was to see whether these findings could be reproduced in a large-scale population-based study. Methods A total of 50,807 (54%) out of 94,194 invited aged ≥20 years or older participated in the third wave of the Nord-Trøndelag Health Study study performed in 2006–2008. Among these, 38,807 (41%) had valid measures of hs-CRP and answered questions on headache and insomnia. Elevated hs-CRP was defined as > 3.0 mg/L. The cross-sectional association with headache was estimated by multivariate analyses using multiple logistic regression. The precision of the odds ratio (OR) was assessed with 95% confidence interval (CI). Results In the fully adjusted model, elevated hs-CRP was associated with migraine (OR 1.14, 95% CI 1.04–1.25) and migraine with aura (OR 1.15, 95% CI 1.03–1.29). The association was strongest among individuals with headache ≥15 days/month for any headache (OR 1.26, 95% CI 1.08–1.48), migraine (OR 1.62, 95% CI 1.21–2.17), and migraine with aura (OR 1.84, 95% CI 1.27–2.67). No clear relationship was found between elevated hs-CRP and headache less than 7 days/month or with insomnia. Conclusions Cross-sectional data from this large-scale population-based study showed that elevated hs-CRP was associated with headache ≥7 days/month, especially evident for migraine with aura.

Download Full-text

Whole-Genome Sequencing for Routine Pathogen Surveillance in Public Health: a Population Snapshot of InvasiveStaphylococcus aureusin Europe

mBio ◽

10.1128/mbio.00444-16 ◽

2016 ◽

Vol 7 (3) ◽

Cited By ~ 123

Author(s):

David M. Aanensen ◽

Edward J. Feil ◽

Matthew T. G. Holden ◽

Janina Dordel ◽

Corin A. Yeats ◽

...

Keyword(s):

Public Health ◽

Whole Genome Sequencing ◽

Genome Sequencing ◽

Large Scale ◽

Bacterial Pathogens ◽

Epidemiological Surveillance ◽

Data Sets ◽

Whole Genome ◽

Bioinformatic Tools ◽

Road Map

ABSTRACTThe implementation of routine whole-genome sequencing (WGS) promises to transform our ability to monitor the emergence and spread of bacterial pathogens. Here we combined WGS data from 308 invasiveStaphylococcus aureusisolates corresponding to a pan-European population snapshot, with epidemiological and resistance data. Geospatial visualization of the data is made possible by a generic software tool designed for public health purposes that is available at the project URL (http://www.microreact.org/project/EkUvg9uY?tt=rc). Our analysis demonstrates that high-risk clones can be identified on the basis of population level properties such as clonal relatedness, abundance, and spatial structuring and by inferring virulence and resistance properties on the basis of gene content. We also show thatin silicopredictions of antibiotic resistance profiles are at least as reliable as phenotypic testing. We argue that this work provides a comprehensive road map illustrating the three vital components for future molecular epidemiological surveillance: (i) large-scale structured surveys, (ii) WGS, and (iii) community-oriented database infrastructure and analysis tools.IMPORTANCEThe spread of antibiotic-resistant bacteria is a public health emergency of global concern, threatening medical intervention at every level of health care delivery. Several recent studies have demonstrated the promise of routine whole-genome sequencing (WGS) of bacterial pathogens for epidemiological surveillance, outbreak detection, and infection control. However, as this technology becomes more widely adopted, the key challenges of generating representative national and international data sets and the development of bioinformatic tools to manage and interpret the data become increasingly pertinent. This study provides a road map for the integration of WGS data into routine pathogen surveillance. We emphasize the importance of large-scale routine surveys to provide the population context for more targeted or localized investigation and the development of open-access bioinformatic tools to provide the means to combine and compare independently generated data with publicly available data sets.

Download Full-text

Adoption of large-scale medical equipment: the impact of competition in the German inpatient sector

The European Journal of Health Economics ◽

10.1007/s10198-021-01395-w ◽

2021 ◽

Author(s):

Marie Dreger ◽

Hauke Langhoff ◽

Cornelia Henschke

Keyword(s):

Medical Equipment ◽

Large Scale ◽

Probability Model ◽

Population Based ◽

Hospital Size ◽

Positron Emission ◽

State Subsidies ◽

Linear Probability ◽

The Impact ◽

Pet Scanners

AbstractThe availability of large-scale medical equipment such as computed tomography (CT), magnet resonance imaging (MRI) and positron emission tomography (PET) scanners has increased rapidly worldwide over the last decades. Among OECD countries, Germany ranks high according to the number of imaging technologies and their applications per inhabitant. In contrast to other countries, there is no active governmental planning of large-scale medical equipment. We therefore investigated whether and how the adoption and distribution of CT, MRI and PET scanners in the German inpatient sector is subject to competition. Using a linear-probability model, we additionally examined the impact of regional, hospital- and population-based factors. In summary, our results indicate that the adoption rate by hospital sites decreases with the number of other sites being already equipped with the respective device and their proximity. However, the effect presumably depends on the technologies’ stage within the diffusion process. No influence regarding the amount of state subsidies could be identified. Furthermore, hospital size and university status strongly affect the adoption.

Download Full-text

The impact of a large-scale population-based asthma management program on pediatric asthma patients and their caregivers

Annals of Allergy Asthma & Immunology ◽

10.1016/s1081-1206(10)61799-1 ◽

2003 ◽

Vol 90 (3) ◽

pp. 308-315 ◽

Cited By ~ 33

Author(s):

Archelle Georgiou ◽

Deborah A. Buchner ◽

Daniel H. Ershoff ◽

Kristin M. Blasko ◽

Linda V. Goodman ◽

...

Keyword(s):

Large Scale ◽

Asthma Management ◽

Pediatric Asthma ◽

Population Based ◽

Management Program ◽

Scale Population ◽

Asthma Patients ◽

The Impact

Download Full-text

Leveraging mobile phone surveys during the COVID-19 pandemic in Ecuador and Sri Lanka: Methods, timeline and findings

PLoS ONE ◽

10.1371/journal.pone.0250171 ◽

2021 ◽

Vol 16 (4) ◽

pp. e0250171

Author(s):

Rachael Phadnis ◽

Champika Wickramasinghe ◽

Juan Carlos Zevallos ◽

Stacy Davlin ◽

Vindya Kumarapeli ◽

...

Keyword(s):

Decision Making ◽

Sri Lanka ◽

Mobile Phone ◽

Technical Assistance ◽

Population Based ◽

Mitigation Strategies ◽

Time Data ◽

Attitudes And Practices ◽

Government Clinic ◽

The Impact

Effective and rapid decision making during a pandemic requires data not only about infections, but also about human behavior. Mobile phone surveys (MPS) offer the opportunity to collect real-time data on behavior, exposure, knowledge, and perception, as well as care and treatment to inform decision making. The surveys aimed to collect coronavirus disease 2019 (COVID-19) related information in Ecuador and Sri Lanka using mobile phones. In Ecuador, a Knowledge, Attitudes and Practices (KAP) survey was conducted. In Sri Lanka, an evaluation of a novel medicine delivery system was conducted. Using the established mobile network operator channels and technical assistance provided through The Bloomberg Philanthropies Data for Health Initiative (D4H), Ministries of Health fielded a population-based COVID-19-specific MPS using Surveda, the open source data collection tool developed as part of the initiative. A total of 1,185 adults in Ecuador completed the MPS in 14 days. A total of 5,001 adults over the age of 35 in Sri Lanka completed the MPS in 44 days. Both samples were adjusted to the 2019 United Nations Population Estimates to produce population-based estimates by age and sex. The Ecuador COVID-19 MPS found that there was compliance with the mitigation strategies implemented in that country. Overall, 96.5% of Ecuadorians reported wearing a face mask or face covering when leaving home. Overall, 3.8% of Sri Lankans used the service to receive medicines from a government clinic. Among those who used the medicine delivery service in Sri Lanka, 95.8% of those who used a private pharmacy received their medications within one week, and 69.9% of those using a government clinic reported the same. These studies demonstrate that MPS can be conducted quickly and gather essential data. MPS can help monitor the impact of interventions and programs, and rapidly identify what works in mitigating the impact of COVID-19.

Download Full-text

Mandrake: visualising microbial population structure by embedding millions of genomes into a low-dimensional representation

10.1101/2021.10.28.466232 ◽

2021 ◽

Author(s):

John A Lees ◽

Gerry Tonkin-Hill ◽

Zhirong Yang ◽

Jukka Corander

Keyword(s):

Population Structure ◽

Large Scale ◽

Population Genomics ◽

Bacterial Species ◽

Population Based ◽

Data Sets ◽

Sequence Alignments ◽

Multiple Sequence ◽

Dimensional Reduction Method ◽

Low Dimensional

In less than a decade, population genomics of microbes has progressed from the effort of sequencing dozens of strains to thousands, or even tens of thousands of strains in a single study. There are now hundreds of thousands of genomes available even for a single bacterial species and the number of genomes is expected to continue to increase at an accelerated pace given the advances in sequencing technology and widespread genomic surveillance initiatives. This explosion of data calls for innovative methods to enable rapid exploration of the structure of a population based on different data modalities, such as multiple sequence alignments, assemblies and estimates of gene content across different genomes. Here we present Mandrake, an efficient implementation of a dimensional reduction method tailored for the needs of large-scale population genomics. Mandrake is capable of visualising population structure from millions of whole genomes and we illustrate its usefulness with several data sets representing major pathogens. Our method is freely available both as an analysis pipeline (https://github.com/johnlees/mandrake) and as a browser-based interactive application (https://gtonkinhill.github.io/mandrake-web/).

Download Full-text

Megacities as drivers of national outbreaks: the role of holiday travel in the spread of infectious diseases

10.1101/737379 ◽

2019 ◽

Cited By ~ 1

Author(s):

Ayesha S. Mahmud ◽

Md. Iqbal Kabir ◽

Kenth Engø-Monsen ◽

Sania Tahmina ◽

Baizid Khoorshid Riaz ◽

...

Keyword(s):

Infectious Diseases ◽

Large Scale ◽

Disease Risk ◽

Human Mobility ◽

Mobile Phone Data ◽

Dhaka City ◽

Population Mobility ◽

Time Data ◽

The Impact

AbstractHuman mobility connects populations and can lead to large fluctuations in population density, both of which are important drivers of epidemics. Measuring population mobility during infectious disease outbreaks is challenging, but is a particularly important goal in the context of rapidly growing and highly connected urban centers in low and middle income countries, which can act to amplify and spread local epidemics nationally and internationally. Here, we combine estimates of population movement from mobile phone data for over 4 million subscribers in the megacity of Dhaka, Bangladesh, one of the most densely populated cities globally. We combine mobility data with epidemiological data from a household survey, to understand the role of population mobility on the spatial spread of the mosquito-borne virus chikungunya within and outside Dhaka city during a large outbreak in 2017. The peak of the 2017 chikungunya outbreak in Dhaka coincided with the annual Eid holidays, during which large numbers of people traveled from Dhaka to their native region in other parts of the country. We show that regular population fluxes around Dhaka city played a significant role in determining disease risk, and that travel during Eid was crucial to the spread of the infection to the rest of the country. Our results highlight the impact of large-scale population movements, for example during holidays, on the spread of infectious diseases. These dynamics are difficult to capture using traditional approaches, and we compare our results to a standard diffusion model, to highlight the value of real-time data from mobile phones for outbreak analysis, forecasting, and surveillance.

Download Full-text

Public policy and economic dynamics of COVID-19 spread: A mathematical modeling study

PLoS ONE ◽

10.1371/journal.pone.0244174 ◽

2020 ◽

Vol 15 (12) ◽

pp. e0244174

Author(s):

Uri Goldsztejn ◽

David Schwartzman ◽

Arye Nehorai

Keyword(s):

Mathematical Modeling ◽

Large Scale ◽

Population Based ◽

Disease Spread ◽

Economic Damage ◽

Baseline Scenario ◽

Economic Output ◽

Economic Vulnerability ◽

Restrictive Measures ◽

The Impact

With the COVID-19 pandemic infecting millions of people, large-scale isolation policies have been enacted across the globe. To assess the impact of isolation measures on deaths, hospitalizations, and economic output, we create a mathematical model to simulate the spread of COVID-19, incorporating effects of restrictive measures and segmenting the population based on health risk and economic vulnerability. Policymakers make isolation policy decisions based on current levels of disease spread and economic damage. For 76 weeks in a population of 330 million, we simulate a baseline scenario leaving strong isolation restrictions in place, rapidly reducing isolation restrictions for non-seniors shortly after outbreak containment, and gradually relaxing isolation restrictions for non-seniors. We use 76 weeks as an approximation of the time at which a vaccine will be available. In the baseline scenario, there are 235,724 deaths and the economy shrinks by 34.0%. With a rapid relaxation, a second outbreak takes place, with 525,558 deaths, and the economy shrinks by 32.3%. With a gradual relaxation, there are 262,917 deaths, and the economy shrinks by 29.8%. We also show that hospitalizations, deaths, and economic output are quite sensitive to disease spread by asymptomatic people. Strict restrictions on seniors with very gradual lifting of isolation for non-seniors results in a limited number of deaths and lesser economic damage. Therefore, we recommend this strategy and measures that reduce non-isolated disease spread to control the pandemic while making isolation economically viable.

Download Full-text

Physical Activity Is Associated With Lower Long-Term Incidence of Anxiety in a Population-Based, Large-Scale Study

Frontiers in Psychiatry ◽

10.3389/fpsyt.2021.714014 ◽

2021 ◽

Vol 12 ◽

Author(s):

Martina Svensson ◽

Lena Brundin ◽

Sophie Erhardt ◽

Ulf Hållmarker ◽

Stefan James ◽

...

Keyword(s):

Physical Activity ◽

Physical Performance ◽

Lower Risk ◽

Large Scale ◽

Population Based ◽

Patient Registries ◽

Swedish Population ◽

Increased Risk ◽

Observational Study Design ◽

The Impact

Physical activity may prevent anxiety, but the importance of exercise intensity, sex-specific mechanisms, and duration of the effects remains largely unknown. We used an observational study design to follow 395,369 individuals for up to 21 years to investigate if participation in an ultralong-distance cross-country ski race (Vasaloppet, up to 90 km) was associated with a lower risk of developing anxiety. Skiers in the race and matched non-skiers from the general population were studied after participation in the race using the Swedish population and patient registries. Skiers (n = 197,685, median age 36 years, 38% women) had a significantly lower risk of developing anxiety during the follow-up compared to non-skiers (adjusted hazard ratio, HR 0.42). However, among women, higher physical performance (measured as the finishing time to complete the race, a proxy for higher exercise dose) was associated with an increased risk of anxiety compared to slower skiing women (HR 2.00). For men, the finishing time of the race did not significantly impact the risk of anxiety. Our results support the recommendations of engaging in physical activity to decrease the risk of anxiety in both men and women. The impact of physical performance level on the risk of anxiety requires further investigations among women.

Download Full-text