Interval estimation for the breakpoint in segmented regression: a smoothed score‐based approach

Muggeo, Vito M. R.

doi:10.1111/anzs.12200

Cited by 239 publications

(184 citation statements)

References 36 publications

Supporting

Mentioning

181

Contrasting

Order By: Relevance

“…Calculations were performed using R version 3.6.1 (R Core Team, 2019). For the entire time series, a segmented Poisson regression model (SEG) allowing for over dispersion (Wood, 2006) was fitted to daily mortality with daily mean temperature as predictor and day-of-week as only covariate using the R package segmented (Muggeo, 2003(Muggeo, , 2008a(Muggeo, , 2017. Estimates and SE were provided for MMT and for the cold and heat slope parameters.…”

Section: Model Calculationsmentioning

confidence: 99%

“…There are several approaches to calculate MMT from temperature-mortality time-series data and different methods have been used in the studies mentioned above. One simple statistical model predicting the logarithm of the death counts by actual temperature is the segmented Poisson regression model (SEG) providing estimates of the breakpoint (MMT) as well as of the negative temperature slope in the cold and positive slope in the heat, while accounting for covariates, e.g., day of week (Muggeo, 2003(Muggeo, , 2008a(Muggeo, , 2017. Only focusing on the temperature influence on the same day, SEG neglects the time series structure, and especially does not consider lagged effects of temperature on mortality.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Long Term Adaptation to Heat Stress: Shifts in the Minimum Mortality Temperature in the Netherlands

et al. 2020

View full text Add to dashboard Cite

It is essentially unknown how humans adapt or will adapt to heat stress caused by climate change over a long-term interval. A possible indicator of adaptation may be the minimum mortality temperature (MMT), which is defined as the mean daily temperature at which the lowest mortality occurs. Another possible indicator may be the heat sensitivity, i.e., the percentage change in mortality per 1 • C above the MMT threshold, or heat attributable fraction (AF), i.e., the percentage relative excess mortality above MMT. We estimated MMT and heat sensitivity/AF over a period of 23 years for older adults (≥65 years) in the Netherlands using three commonly used methods. These methods are segmented Poisson regression (SEG), constrained segmented distributed lag models (CSDL), and distributed lag non-linear models (DLNM). The mean ambient temperature increased by 0.03 • C/year over the 23 year period. The calculated mean MMT over the 23-year period differed considerably between methods [16.4 ± 1.2 • C (SE) (SEG), 18.9 ± 0.5 • C (CSDL), and 15.3 ± 0.4 • C DLNM]. MMT increased during the observed period according to CSDL (0.11 ± 0.05 • C/year) and DLNM (0.15 ± 0.02 • C/year), but not with SEG. The heat sensitivity, however, decreased for the latter method (0.06%/ • C/year) and did not change for CSDL. Heat AF was calculated for the DLNM method and decreased with 0.07%/year. Based on these results we conclude that the susceptibility of humans to heat decreases over time, regardless which method was used, because human adaptation is shown by either an increase in MMT (CSDL and DLNM) or a decrease in heat sensitivity for unchanged MMT (SEG). Future studies should focus on what factors (e.g., physiological, behavioral, technological, or infrastructural adaptations) influence human adaptation the most, so it can be promoted through adaptation policies. Furthermore, future studies should keep in mind that the employed method influences the calculated MMT, which hampers comparability between studies.

show abstract

Section: Model Calculationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Long Term Adaptation to Heat Stress: Shifts in the Minimum Mortality Temperature in the Netherlands

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Using R, log likelihood was calculated for a linear regression model and a segmented threshold model from 'segmented' package (Muggeo, 2017), P value was calculated using the likelihood ratio test (LRT) with a χ2 distribution. For comparison, the set of all 519 JASPAR motifs, or the 349 JASPR motifs corresponding to ESC expressed TFs were used to identify the number of TFBS in each sequence and the relationship to flanking H3K27ac ( Figure S4).…”

Section: Tfbs Conservation and Enrichment Analysismentioning

confidence: 99%

A flexible repertoire of transcription factor binding sites and diversity threshold determines enhancer activity in embryonic stem cells

Singh

Mullany

Moorthy

et al. 2020

Preprint

View full text Add to dashboard Cite

Transcriptional enhancers are critical for development, phenotype evolution and often mutated in disease contexts; however, even in well-studied cell types, the sequence code conferring enhancer activity remains unknown. We found genomic regions with conserved binding of multiple transcription factors (TFs) in mouse and human embryonic stem cells (ESCs) are enriched in a diverse repertoire of transcription factor binding sites (TFBS) including known and novel ESC regulators. Remarkably, using a diverse set of TFBS from this repertoire was sufficient to construct short synthetic enhancers with activity comparable to native enhancers. Site directed mutagenesis of conserved TFBS in endogenous enhancers or TFBS deletion from synthetic sequences revealed a requirement for >10 different TFBS for robust activity. Specific TFBS, including the OCT4:SOX2 co-motif, are dispensable, despite co-binding the OCT4, SOX2 and NANOG master regulators of pluripotency. These findings reveal a TFBS diversity threshold overrides the need for optimized regulatory grammar and individual TFBS that bind specific master regulators. Highlights: Comparative epigenomics determines the enhancer sequence code and synthetic enhancer design. A diversity of >10 TFBS are required and sufficient for robust enhancer activity  >40 TFBS contribute to enhancer activity implicating new TFs in pluripotency maintenance  Increased TFBS diversity, above a threshold, overrides the need for regulatory grammar

show abstract

“…before or after the observed break in species prevalence through time. The break in species prevalence occurring around the year 2000 was inferred after regression species prevalence upon the year, using the segmented package (Muggeo, 2017). This strategy was chosen to account for the temporal trends in species prevalence and abundance; more complex mixed models including year as a random effect did not provide precise year effect estimates and were faced with convergence issues when combined with a negative binomial link function.…”

Section: Methodsmentioning

confidence: 99%

Compilation of 29-yearpostmortemexaminations identifies major shifts in equine parasite prevalence from 2000 onwards

Sallé

Guillot

Tapprest

et al. 2019

Preprint

View full text Add to dashboard Cite

0Horses are infected by a wide range of parasite species that form complex communities. 1Parasite control imposes significant constraints on parasite species assemblage whose 2 2 monitoring remains however difficult to track through time. Postmortem examination is a 2 3 reliable method to quantify parasite communities. Here, we compiled 1,673 necropsy reports 2 4 accumulated over 29 years, in the reference necropsy centre from Normandy (France). The 2 5 burden of non-strongylid species was quantified. Details of horse deworming history and the 2 6 cause of death (resulting from a parasitic infection or not) were registered. Data analyses 2 7 revealed the seasonal fluctuations of non-strongylid parasite species and the least exposure 2 8of race horses to these parasites. Beyond these observations, we found a shift in the species 2 9 responsible for fatal parasitic infection from the year 2000 onward, whereby fatal 3 0 cyathostominosis and Parascaris spp. infection have replaced death cases caused by S. 3 1 vulgaris and tapeworms. Concomitant break in the temporal trend of parasite species 3 2prevalence was also found within a 10-year window (1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)) that has seen the rise of 3 3Parascaris spp. and the decline of both Gasterophilus spp. and tapeworms. A few cases of 3 4 parasite persistence following deworming were identified that all occurred after 2000.3 5Altogether, these findings provide a unique assessment of the major remodelling of equine 3 6 parasitic communities over the last 29 years. They also underscore the critical importance of 3 7Parascaris spp. in young equids.3 8 3 9

show abstract

Interval estimation for the breakpoint in segmented regression: a smoothed score‐based approach

Cited by 239 publications

References 36 publications

Long Term Adaptation to Heat Stress: Shifts in the Minimum Mortality Temperature in the Netherlands

Long Term Adaptation to Heat Stress: Shifts in the Minimum Mortality Temperature in the Netherlands

A flexible repertoire of transcription factor binding sites and diversity threshold determines enhancer activity in embryonic stem cells

Compilation of 29-yearpostmortemexaminations identifies major shifts in equine parasite prevalence from 2000 onwards

Contact Info

Product

Resources

About