Search

Humans play an integral role in identifying important information from social media during disasters. While human annotation of social media data to train machine learning models is often viewed as human-computer interaction, this study interrogates the ontological boundary between such interaction and human-machine communication. We conducted multiple interviews with participants who both labeled data to train machine learning models and corrected machine-inferred data labels. Findings reveal three themes: scripts invoked to manage decision-making, contextual scripts, and scripts around perceptions of machines. Humans use scripts around training the machine—a form of behavioral anthropomorphism—to develop social relationships with them. Correcting machine-inferred data labels changes these scripts and evokes self-doubt around who is right, which substantiates the argument that this is a form of human-machine communication.

Luciferase-based reporters provide a key measurement approach in a broad range of applications, from in vitro high-throughput screening to whole animal imaging. For example, luminescence intensity is widely used to measure promoter activity, protein expression levels, and cell growth. However, luminescence intensity measurements are subject to quantitative irregularities caused by luminescence decay and variation in reporter expression level. In contrast, bioluminescence resonance energy transfer (BRET) sensors provide the advantages of luciferase-based reporters but overcome the aforementioned irregularities because of the inherently ratiometric readout. Here, we generated a new ratiometric BRET sensor of ATP (ARSeNL—ATP detection with a Ratiometric mScarlet-NanoLuc sensor), and we demonstrated that it provides a stable and robust readout across protein, cell, and whole animal tissue contexts. The ARSeNL sensor was engineered by screening a color palette of sensors utilizing variants of the high photon flux NanoLuc luciferase as donors and a panel of red fluorescent proteins as acceptors. We found that the novel combination of NanoLuc and mScarlet exhibited the largest dynamic range, with a 5-fold change in the BRET ratio upon saturation with ATP. Importantly, the NanoLuc-mScarlet BRET pair provided a large spectral separation between luminescence emission channels that is compatible with green and red filter sets extensively used in typical biological microscopes and animal imaging systems. Using this new sensor, we showed that the BRET ratio was independent of luminescence intensity decay and sensor expression level, and the BRET ratio faithfully reported differences in live-cell energy metabolism whether in culture or within mouse tissue. In particular, BRET analyte sensors have not been used broadly in tissue contexts, and thus, in principle, our sensor could provide a new tool for in vivo imaging of metabolic status.

Purinergic signals, such as extracellular adenosine triphosphate (ATP) and adenosine diphosphate (ADP), mediate intercellular communication and stress responses throughout mammalian tissues, but the dynamics of their release and clearance are still not well understood. Although physiochemical methods provide important insight into physiology, genetically encoded optical sensors have proven particularly powerful in the quantification of signaling in live specimens. Indeed, genetically encoded luminescent and fluorescent sensors provide new insights into ATP-mediated purinergic signaling. However, new tools to detect extracellular ADP are still required. To this end, in this study, we use protein engineering to generate a new genetically encoded sensor that employs a high-affinity bacterial ADP-binding protein and reports a change in occupancy with a change in the Förster-type resonance energy transfer (FRET) between cyan and yellow fluorescent proteins. We characterize the sensor in both protein solution studies, as well as live-cell microscopy. This new sensor responds to nanomolar and micromolar concentrations of ADP and ATP in solution, respectively, and in principle it is the first fully-genetically encoded sensor with sufficiently high affinity for ADP to detect low levels of extracellular ADP. Furthermore, we demonstrate that tethering the sensor to the cell surface enables the detection of physiologically relevant nucleotide release induced by hypoosmotic shock as a model of tissue edema. Thus, we provide a new tool to study purinergic signaling that can be used across genetically tractable model systems.

Abstract Hailstorms are analyzed across the United States using explicit hailstone size calculations from convection-permitting regional climate simulations for historical, mid-century, and end of twenty-first-century epochs. Near-surface hailstones

Abstract Mesoscale convective systems (MCSs) are a substantial source of precipitation in the eastern U.S. and may be sensitive to regional climatic change. We use a suite of convection-permitting climate simulations to examine possible changes in MCS precipitation. Specifically, annual and regional totals of MCS and non-MCS precipitation generated during a retrospective simulation are compared to end-of-21st-century simulations based on intermediate and extreme climate change scenarios. Both scenarios produce more MCS precipitation and less non-MCS precipitation, thus significantly increasing the proportion of precipitation associated with MCSs across the U.S.

Abstract Explicit representation of finer‐scale processes can affect the sign and magnitude of the precipitation response to climate change between convection‐permitting and convection‐parameterizing models. We compare precipitation across two 15‐year epochs, a historical (HIST) and an end‐of‐21st‐century (EoC85), between a set of dynamically downscaled regional climate simulations at 3.75 km grid spacing (WRF) and bias‐corrected Community Earth System Model (CESM) output used to initialize and force the lateral boundaries of the downscaled simulations. In the historical climate, the downscaled simulations demonstrate less overall error than CESM when compared to observations for most portions of the conterminous United States. Both sets of simulations overestimate the incidence of environments with moderate to high precipitable water while CESM generally simulates rainfall that is too frequent but less intense. Within both sets of simulations, EoC85 rainfall amounts decrease in low‐moisture environments due to reduced rainfall frequency and intensity while rainfall amounts increase in high‐moisture environments as they occur more often. Overall, reductions in rainfall are stronger in WRF than in CESM, particularly during the warm season. This reduced drying in CESM is attributed to relatively higher rainfall frequency in environments with high concentrations of precipitable water and weak vertical motion. As a result, an increase in the occurrence of high moisture environments in EoC85 naturally favors more rainfall in CESM than WRF. Our results present an in‐depth examination of the characteristics of changes in overall accumulated precipitation and highlight an extra dimension of uncertainty when comparing convection‐permitting models against convection‐parameterizing models.

Background: The validated Arthroplasty Risk Score (ARS) predicts the need for postoperative triage to an intensive care setting. We hypothesized that the ARS may also predict hospital length of stay (LOS), discharge disposition, and episode-of-care cost (EOCC). Methods: We retrospectively reviewed a series of 704 patients undergoing primary total hip and knee arthroplasty over 17 months. Patient characteristics, 90-day EOCC, LOS, and readmission rates were compared before and after ARS implementation. Results: ARS implementation was associated with fewer patients going to a skilled nursing or rehabilitation facility after discharge (63% vs 74%, P = .002). There was no difference in LOS, EOCC, readmission rates, or complications. While the adoption of the ARS did not change the mean EOCC, ARS >3 was predictive of high EOCC outlier (odds ratio 2.65, 95% confidence interval 1.40-5.01, P = .003). Increased ARS correlated with increased EOCC (P = .003). Conclusions: Implementation of the ARS was associated with increased disposition to home. It was predictive of high EOCC and should be considered in risk adjustment variables in alternative payment models. Keywords: Bundled payments, Risk stratification, Arthroplasty