Search

Start Over Source dataspace

61. Fast animal pose estimation using deep neural networks

Author(s):
Pereira, Talmo D.; Aldarondo, Diego E.; Willmore, Lindsay; Kislin, Mikhail; Wang, Samuel S.-H.; Murthy, Mala; Shaevitz, Joshua W.
Abstract:
Recent work quantifying postural dynamics has attempted to define the repertoire of behaviors performed by an animal. However, a major drawback to these techniques has been their reliance on dimensionality reduction of images which destroys information about which parts of the body are used in each behavior. To address this issue, we introduce a deep learning-based method for pose estimation, LEAP (LEAP Estimates Animal Pose). LEAP automatically predicts the positions of animal body parts using a deep convolutional neural network with as little as 10 frames of labeled data for training. This framework consists of a graphical interface for interactive labeling of body parts and software for training the network and fast prediction on new data (1 hr to train, 185 Hz predictions). We validate LEAP using videos of freely behaving fruit flies (Drosophila melanogaster) and track 32 distinct points on the body to fully describe the pose of the head, body, wings, and legs with an error rate of <3% of the animal's body length. We recapitulate a number of reported findings on insect gait dynamics and show LEAP's applicability as the first step in unsupervised behavioral classification. Finally, we extend the method to more challenging imaging situations (pairs of flies moving on a mesh-like background) and movies from freely moving mice (Mus musculus) where we track the full conformation of the head, body, and limbs.
Type:
Dataset
Issue Date:
30 May 2018

62. Data for Nature Climate Change article 'Regional dry-season climate changes due to three decades of Amazonian deforestation'

Author(s):
Khanna, Jaya; Medvigy, David; Fueglistaler, Stephan; Walko, Robert
Abstract:
More than 20% Amazon rainforest has been cleared in the past three decades triggering important hydroclimatic changes. Small-scale (~few kilometers) deforestation in the 1980s has caused thermally-triggered atmospheric circulations that increase regional cloudiness and precipitation frequency. However, these circulations are predicted to diminish as deforestation increases. Here we use multi-decadal satellite records and numerical model simulations to show a regime shift in the regional hydroclimate accompanying increasing deforestation in Rondônia, Brazil. Compared to the 1980s, present-day deforested areas in downwind western Rondônia are found to be wetter than upwind eastern deforested areas during the local dry season. The resultant precipitation change in the two regions is approximately ±25% of the deforested area mean. Meso-resolution simulations robustly reproduce this transition when forced with increasing deforestation alone, showing a negligible role of large-scale climate variability. Furthermore, deforestation-induced surface roughness reduction is found to play an essential role in the present-day dry season hydroclimate. Our study illustrates the strong scale-sensitivity of the climatic response to Amazonian deforestation and suggests that deforestation is sufficiently advanced to have caused a shift from a thermally- to a dynamically-driven hydroclimatic regime.
Type:
Dataset and Software
Issue Date:
2017

65. Dynamic reconfiguration of the default mode network during narrative comprehension

Author(s):
Simony, Erez; Honey, Christopher; Chen, Janice; Lositsky, Olga; Yeshurun, Yaara; Wiesel, Ami; Hasson, Uri
Abstract:
Does the default mode network (DMN) reconfigure to encode information about the changing environment? This question has proven difficult, because patterns of functional connectivity reflect a mixture of stimulus-induced neural processes, intrinsic neural processes and non-neuronal noise. Here we introduce inter-subject functional correlation (ISFC), which isolates stimulus-dependent inter-regional correlations between brains exposed to the same stimulus. During fMRI, we had subjects listen to a real-life auditory narrative and to temporally scrambled versions of the narrative. We used ISFC to isolate correlation patterns within the DMN that were locked to the processing of each narrative segment and specific to its meaning within the narrative context. The momentary configurations of DMN ISFC were highly replicable across groups. Moreover, DMN coupling strength predicted memory of narrative segments. Thus, ISFC opens new avenues for linking brain network dynamics to stimulus features and behaviour.
Type:
Dataset
Issue Date:
18 July 2016

66. Attentional Modulation of Brain Responses to Primary Appetitive and Aversive Stimuli

Author(s):
Cara L. Buck; Jonathan D. Cohen; Field, Brent; Daniel Kahneman; Samuel M. McClure; Leigh E. Nystrom
Abstract:
Studies of subjective well-being have conventionally relied upon self-report, which directs subjects’ attention to their emotional experiences. This method presumes that attention itself does not influence emotional processes, which could bias sampling. We tested whether attention influences experienced utility (the moment-by-moment experience of pleasure) by using functional magnetic resonance imaging (fMRI) to measure the activity of brain systems thought to represent hedonic value while manipulating attentional load. Subjects received appetitive or aversive solutions orally while alternatively executing a low or high attentional load task. Brain regions associated with hedonic processing, including the ventral striatum, showed a response to both juice and quinine. This response decreased during the high-load task relative to the low-load task. Thus, attentional allocation may influence experienced utility by modulating (either directly or indirectly) the activity of brain mechanisms thought to represent hedonic value.
Type:
Dataset, Software, and text
Issue Date:
11 February 2015

67. Basin-Scale Leakage Risks from Geologic Carbon Sequestration: Impact on Carbon Capture and Storage Energy Market Competitiveness

Abstract:
This three-year project, performed by Princeton University in partnership with the University of Minnesota and Brookhaven National Laboratory, examined geologic carbon sequestration in regard to CO2 leakage and potential subsurface liabilities. The research resulted in basin-scale analyses of CO2 and brine leakage in light of uncertainties in the characteristics of leakage processes, and generated frameworks to monetize the risks of leakage interference with competing subsurface resources. The geographic focus was the Michigan sedimentary basin, for which a 3D topographical model was constructed to represent the hydrostratigraphy. Specifically for Ottawa County, a statistical analysis of the hydraulic properties of underlying sedimentary formations was conducted. For plausible scenarios of injection into the Mt. Simon sandstone, leakage rates were estimated and fluxes into shallow drinking-water aquifers were found to be less than natural analogs of CO2 fluxes. We developed the Leakage Impact Valuation (LIV) model in which we identified stakeholders and estimated costs associated with leakage events. It was found that costs could be incurred even in the absence of legal action or other subsurface interference because there are substantial costs of finding and fixing the leak and from injection interruption. We developed a model framework called RISCS, which can be used to predict monetized risk of interference with subsurface resources by combining basin-scale leakage predictions with the LIV method. The project has also developed a cost calculator called the Economic and Policy Drivers Module (EPDM), which comprehensively calculates the costs of carbon sequestration and leakage, and can be used to examine major drivers for subsurface leakage liabilities in relation to specific injection scenarios and leakage events. Finally, we examined the competitiveness of CCS in the energy market. This analysis, though qualitative, shows that financial incentives, such as a carbon tax, are needed for coal combustion with CCS to gain market share. In another part of the project we studied the role of geochemical reactions in affecting the probability of CO2 leakage. A basin-scale simulation tool was modified to account for changes in leakage rates due to permeability alterations, based on simplified mathematical rules for the important geochemical reactions between acidified brines and caprock minerals. In studies of reactive flows in fractured caprocks, we examined the potential for permeability increases, and the extent to which existing reactive transport models would or would not be able to predict it. Using caprock specimens from the Eau Claire and Amherstburg, we found that substantial increases in permeability are possible for caprocks that have significant carbonate content, but minimal alteration is expected otherwise. We also found that while the permeability increase may be substantial, it is much less than what would be predicted from hydrodynamic models based on mechanical aperture alone because the roughness that is generated tends to inhibit flow.
Type:
text
Issue Date:
5 June 2014

68. Bubble pinch-off in turbulence

Abstract:
Experimental data, simulation code, and Python scripts to reproduce the data presented in "Bubble pinch-off in turbulence"
Type:
Dataset and Software

69. Instability of shear flows in spatially periodic domains

Abstract:
Force-driven parallel shear flow in a spatially periodic domain is shown to be linearly unstable with respect to both the Reynolds number and the domain aspect ratio. This finding is confirmed by computer simulations, and a simple expression is derived to determine stable flow conditions. Periodic extensions of Couette and Poiseuille flows are unstable at Reynolds numbers two orders of magnitude smaller than their aperiodic equivalents because the periodic boundaries impose fundamentally different constraints. This instability has important implications for designing computational models of nonlinear dynamic processes with periodicity.
Type:
Dataset

70. Regulation of Harvester Ant Foraging as a Closed-Loop Excitable System

Abstract:
Ant colonies regulate activity in response to changing conditions without using centralized control. Harvester ant colonies forage in the desert for seeds, and their regulation of foraging manages a tradeoff between spending and obtaining water. Foragers lose water while outside in the dry air, but the colony obtains water by metabolizing the fats in the seeds they eat. Previous work shows that the rate at which an outgoing forager leaves the nest depends on its recent experience of brief antennal contact with returning foragers that carry a seed. We examine how this process can yield foraging rates that are robust to uncertainty and responsive to temperature and humidity across minutes to hour-long timescales. To explore possible mechanisms, we develop a low-dimensional analytical model with a small number of parameters that captures observed foraging behavior. The model uses excitability dynamics to represent response to interactions inside the nest and a random delay distribution to represent foraging time outside the nest. We show how feedback of outgoing foragers returning to the nest stabilizes the incoming and outgoing foraging rates to a common value determined by the ``volatility’’ of available foragers. The model exhibits a critical volatility above which there is sustained foraging at a constant rate and below which there is cessation of foraging. To explain how the foraging rates of colonies adjust to temperature and humidity, we propose a mechanism that relies on foragers modifying their volatility after they leave the nest and get exposed to the environment. Our study highlights the importance of feedback in the regulation of foraging activity and points to modulation of volatility as a key to explaining differences in foraging activity in response to conditions and across colonies. Our results present opportunities for generalization to other contexts and systems with excitability and feedback across multiple timescales.
Type:
collection and Dataset