Using Piecewise Regression to Identify Biological Phenomena in Biotelemetry Datasets

Technological advances in the field of animal tracking have greatly expanded the potential to remotely monitor animals, opening the door to exploring how animals shift their behavior over time or respond to external stimuli. A wide variety of animal-borne sensors can provide information on an animal's location, movement characteristics, external environmental conditions, and internal physiological status. Here, we demonstrate how piecewise regression can be used to identify the presence and timing of potential shifts in a variety of biological responses using GPS telemetry and other biologging data streams. Different biological latent states can be inferred by partitioning a time-series into multiple segments based on changes in modeled responses (e.g., their mean, variance, trend, degree of autocorrelation) and specifying a unique model structure for each interval. We provide five example applications highlighting a variety of taxonomic species, data streams, timescales, and biological phenomena. These examples include a short-term behavioral response (flee and return) by a trumpeter swan (Cygnus buccinator) immediately following a GPS collar deployment; remote identification of parturition based on movements by a pregnant moose (Alces alces); a physiological response (spike in heart-rate) in a black bear (Ursus americanus) to a stressful stimulus (presence of a drone); a mortality event of a trumpeter swan signaled by changes in collar temperature and Overall Dynamic Body Acceleration; and an unsupervised method for identifying the onset, return, duration, and staging use of sandhill crane (Antigone canadensis) migration. We implement analyses using the mcp package in R, which provides functionality for specifying and fitting a wide variety of user-defined model structures in a Bayesian framework and methods for assessing and comparing models using information criterion and cross-validation measures. This approach uses simple modeling approaches that are accessible to a wide audience and is a straightforward means of assessing a variety of biologically relevant changes in animal behavior.

Detecting the Turning Points of Grassland Autumn Phenology on the Qinghai-Tibetan Plateau: Spatial Heterogeneity and Controls

Autumn phenology, commonly represented by the end of season (EOS), is considered to be the most sensitive and crucial productivity indicator of alpine and cold grassland in the Qinghai-Tibetan Plateau. Previous studies typically assumed that the rates of EOS changes remain unchanged over long time periods. However, pixel-scale analysis indicates the existence of turning points and differing EOS change rates before and after these points. The spatial heterogeneity and controls of these turning points remain unclear. In this study, the EOS turning point changes are extracted and their controls are explored by integrating long time-series remote sensing images and piecewise regression methods. The results indicate that the EOS changed over time with a delay rate of 0.08 days/year during 1982–2015. The rates of change are not consistent over different time periods, which clearly highlights the existence of turning points. The results show that temperature contributed most strongly to the EOS changes, followed by precipitation and insolation. Furthermore, the turning points of climate, human activities (e.g., grazing, economic development), and their intersections are found to jointly control the EOS turning points. This study is the first quantitative investigation into the spatial heterogeneity and controls of the EOS turning points on the Qinghai-Tibetan Plateau, and provides important insight into the growth mechanism of alpine and cold grassland.

Metop First Generation AVHRR FRAC SST Reanalysis Version 1

The first full-mission global AVHRR FRAC sea surface temperature (SST) dataset with a nominal 1.1km resolution at nadir was produced from three Metop First Generation (FG) satellites: Metop-A (2006-on), -B (2012-on) and -C (2018-on), using the NOAA Advanced Clear Sky Processor for Ocean (ACSPO) SST enterprise system. Historical reprocessing (‘Reanalysis-1’, RAN1) starts at the beginning of each mission and continues into near-real time (NRT). ACSPO generates two SST products, one with global regression (GR; highly sensitive to skin SST), and another one with piecewise regression (PWR; proxy for depth SST) algorithms. Small residual effects of orbital and sensor instabilities on SST retrievals are mitigated by retraining the regression coefficients daily, using matchups with drifting and tropical moored buoys within moving time windows. In RAN, the training windows are centered at the processed day. In NRT, the same size windows are employed but delayed in time, ending four to ten days prior to the processed day. Delayed-mode RAN reprocessing follows the NRT with a two-month lag, resulting in a higher quality and a more consistent SST record. In addition to its completeness, the newly created Metop-FG RAN1 SST dataset shows very close agreement with in situ data (including the fully independent Argo floats), well within the NOAA specifications for accuracy (global mean bias; ±0.2 K) and precision (global standard deviation; 0.6 K) in a ~20% clear-sky domain (percent of clear-sky SST pixels to the total of ice-free ocean). All performance statistics are stable in time, and consistent across the three platforms. The Metop-FG RAN1 data set is archived at the NASA JPL PO.DAAC and NOAA NCEI. This paper documents the newly created dataset and evaluates its performance.

Quantifying the Onset of Unintended Weight Loss in Huntington’s Disease: A Retrospective Analysis of Enroll-HD

Background: Unintended weight loss and decreased body mass indexes (BMIs) are common symptoms of individuals with manifest HD. It is unknown at what point during disease progression weight loss starts to accelerate relative to a healthy individual’s weight and when recommended interventions should be initiated to have the strongest impact on patient care. Objective: The objective of this study was to identify a point in time relative to age at motor onset when the decline in weight in HD starts to accelerate relative to a non-HD population. The relationship between initiation of weight loss interventions and changes in weight loss was also explored. Methods: Participants from the fifth version of the Enroll-HD study were identified for this research. Linear mixed-effects piecewise regression models were used to estimate the point in time relative to the reported age of motor onset in which BMI started to decline in participants with HD compared to healthy non-HD controls. A post-hoc descriptive analysis was performed to look at when nutritional supplements and swallow therapy were initiated in participants with HD relative to motor onset. Results: BMI decline in the HD group began to accelerate compared to controls approximately 5.7 years after the reported age of motor onset (95% CI: 4.7–6.9). The average initiation times of swallow therapy and nutritional supplements were 7.7 years (SD = 5.5 years) and 6.7 years (SD = 6.5 years) after motor onset, respectively. Conclusion: Our findings suggest a potential point for intervention of nutrition programs or therapies used to prevent future weight loss.

A Bayesian Analysis of the Inversion of the SARS-COV-2 Case Rate in the Countries of the 2020 European Football Championship

While Europe was beginning to deal with the resurgence of COVID-19 due to the Delta variant, the European football championship took place from 11 June to 11 July 2021. We studied the inversion in the decreased/increased rate of new SARS-COV-2 infections in the countries of the tournament, investigating the hypothesis of an association. Using a Bayesian piecewise regression with a Poisson generalized linear model, we looked for a changepoint in the timeseries of the new SARS-COV-2 cases of each country, expecting it to appear not later than two to three weeks after the date of their first match. The two slopes, before and after the changepoint, were used to discuss the reversal from a decreasing to an increasing rate of the infections. For 17 out of 22 countries (77%) the changepoint came on average 14.97 days after their first match (95% CI 12.29–17.47). For all those 17 countries, the changepoint coincides with an inversion from a decreasing to an increasing rate of the infections. Before the changepoint, the new cases were decreasing, halving on average every 18.07 days (95% CI 11.81–29.42). After the changepoint, the cases begin to increase, doubling every 29.10 days (95% CI 14.12–9.78). This inversion in the SARS-COV-2 case rate, which happened during the tournament, provides evidence in favor of a relationship.

A Bayesian Analysis of the Inversion of the SARS-COV-2 Case Rate in the Countries of the 2020 European Football Championship

While Europe was beginning to deal with the resurgence of COVID-19 due to the Delta variant, the European football championship took place, June 11 - July 11, 2021. We studied the inversion in the decrease/increase rate of new SARS-COV-2 infections in the countries of the tournament, investigating the hypothesis of an association. Using a Bayesian piecewise regression with a Poisson Generalized Linear Model, we looked for a changepoint in the timeseries of the new SARS-COV-2 cases of each country, expecting it to appear not later than two to three weeks after the date of their first match. The two slopes, before and after the changepoint, were used to discuss the reversal from a decreasing to an increasing rate of the infections. For 17 out of 22 countries (77%) the changepoint came on average 14.97 days after their first match [95% CI 12.29 to 17.47]. For all those 17 countries, the changepoint coincides with an inversion from a decreasing to an increasing rate of the infections. Before the changepoint, the new cases were decreasing, halving on average every 18.07 days [95% CI 11.81 to 29.42]. After the changepoint, the cases begin to increase, doubling every 29.10 days [95% CI 14.12 to 49.78]. This inversion in the SARS-COV-2 case rate, happened during the tournament, provides evidence in favor of a relationship

A Bayesian Analysis of the Inversion of the SARS-COV-2 Case Rate in the Countries of the 2020 European Football Championship

While Europe was beginning to deal with the resurgence of COVID-19 due to the Delta variant, the European football championship took place, June 11 - July 11, 2021. We studied the inversion in the decrease/increase rate of new SARS-COV-2 infections in the countries of the tournament, investigating the hypothesis of an association. Using a Bayesian piecewise regression with a Poisson Generalized Linear Model, we looked for a changepoint in the timeseries of the new SARS-COV-2 cases of each country, expecting it to appear within four weeks since the date of their first match. The two slopes, before and after the changepoint, were used to discuss the reversal from a decreasing to an increasing rate of the infections. 17 out of 22 countries (77%) have had a changepoint 14.97 days after their first match [95% CI 12.29 to 17.47]. For all those 17 countries, the changepoint coincides with an inversion from a decreasing to an increasing rate of the infections. Before the changepoint, the new cases were decreasing, halving on average every 18.07 days [95% CI 11.81 to 29.42]. After the changepoint, the cases begin to increase, doubling every 29.10 days [95% CI 14.12 to 49.78]. This inversion in the SARS-COV-2 case rate, happened during the tournament, provides evidence in favor of a relationship.

MO188RECONSIDERING THE EDELMAN EQUATION: IMPACT OF INDIVIDUAL TOTAL BODY CATION CONTENT AND BODY WEIGHT

Background and Aims Treatment of dysnatremias is guided by formulas that are based on the Edelman equation, including Adrogue-Madias’ and others. Edelman’s equation is the result of a unique study in which serum sodium concentration ([Na+]), total body exchangeable sodium (Nae+) and potassium (Ke+) and total body water (TBW) were measured in a highly heterogeneous population. Because these observations resulted from steady state observations, the equation might not account for the recently uncovered highly dynamic Na+ body compartment where Na+ is temporarily stored and released without affecting TBW. Various factors that influence this Na+ body compartment have been identified and we questioned as to what extent these factors affect associations between serum [Na+] and [Nae+Ke]/TBW. Method We performed a post-hoc analysis of original data published by Edelman. In the linear regression model, effects of sex, edema (y/n), age and body weight (>/< median split) were examined. Serum [Na+] was calculated by multiplying serum water [Na+] of the original data set by 0.93. Using piecewise regression, we analyzed differences in slope and y-intercept for the regression between serum [Na+] and increasing values of (Nae+Ke)/TBW, in which the clinical characteristics from the subgroups were included as interaction factors . Results Data was available for 85 measurements in 82 patients; 57 males and 25 females. The median age (range) age was 58 (27-90) years and median weight (range) was 59.6 (36.4 – 168.2) kilograms. Median serum [Na+] (range) was 131.4 (103.4 – 150.2) mmol/L. The association between serum [Na+] and (Nae+Ke)/TBW was different for high and low weight categories (figure 1A). Sex, age or presence of edema did not alter the relationship. Piecewise regression showed a significant decrease in slope in the regression between serum [Na+] and (Nae+Ke)/TBW above 149 mmol/L (Nae+Ke)/TBW (figure 1B). Conclusion Edelman equation’s coefficients are significantly affected by weight and total body cation content. In subjects with a low weight and low total body cation content, the Edelman equation seems to adequately predict the course of serum [Na+]. However, the clinical use of the Edelman-based formulas may be hampered in subjects with higher weight and higher total body cation content, which may reflect increased tissue Na+ storage. Our analysis underlines the importance of further research into the role of osmotically inactive Na+ storage in osmoregulation.