Yale University held a poster session during the afternoon of December 1 from 4-6 PM as part of the Yale Day of Data. Digital versions of the posters will be posted to the EliScholar page in January.


Subscribe to RSS Feed

Closing the Water Budget in an Experimental Urban Watershed: A Comparative Assessment of Methods for Measuring Evapotranspiration

Leana M. Weissberg, Yale University
Gaboury Benoit, Yale University

Data Collection and Analysis at the ATLAS Detector

Savannah Thais, Yale University

Do elephants eat more trees when less grass is available? A field study in Kruger National Park, South Africa.

Emily Goldberg, Yale

Although African bush elephants are often responsible for the majority of herbivore-driven savanna tree mortality, confusion remains regarding the factors that influence their diet. Some elephants either browse or graze almost exclusively, while others balance the two, and the determinants of this variation remain poorly understood. I seek to determine whether grass availability controls the proportion of woody plants in elephant diet and, therefore, the amount of damage elephants do to trees while foraging. Preliminary analysis using already-available data on grass biomass, elephant density, and elephant damage to trees suggest that tree damage is in fact negatively correlated with grass availability. However, there is a negative correlation between grass biomass and elephant density and a positive one between elephant density and tree damage, which may be sufficient to explain this pattern. This finding, while interesting, does not therefore answer the question of the direct effect of grass availability on tree damage. To resolve this issue, I intend to use my own data to separate the effect of population density from that of grass availability. I gathered data in Kruger National Park, South Africa, on elephant diet composition (by taking dung samples that I will analyze for carbon isotope ratios) and grass biomass (using a disk pasture meter). I will perform the required isotope analyses and evaluate the combination of datasets over the course of this fall.

Extracting Geography From Datasets in Social Sciences

Yuke Li
Tianhao Wu, Yale University
Nicholas Marshall, Yale University
Stefan Steinerberger, Yale University

Machine learning to predict pupillary dynamics in conscious visual perception

Jacob S. Prince, Yale University
Jackson Ding, Yale University
Owen Morgan, St. John's College

Many recent studies have attempted to isolate the neural correlates of consciousness. A promising paradigm involves contrasting the mechanisms involved in the conscious perception of a visual stimulus with those at work when the same stimulus is presented but not consciously seen by the subject. One difficulty these studies often present is that they rely on the subject’s report (usually via button presses) to gauge his or her conscious awareness of the stimulus––the act of reporting upon one’s experience likely induces extra cognitive activity beyond what naturally occurs during conscious perception, such as memorizing some feature of the stimulus or executing a button press. To address this fundamental issue, we investigated whether pupillary responses could be used as a reliable covert indicator of conscious perception during a visual stimulus detection task. Pupil dilation has been previously observed during perceptual switching in binocular rivalry, and at the moment of object or facial recognition. We measured pupil diameter, gaze location, and eye-blinks continuously as subjects were presented with facial stimuli, calibrated to a low opacity such that it would be perceived during approximately 50% of trials. Significant diameter changes were observed in response to faces that were perceived by participants––on average, a peak dilation response occurred between 1 and 2 seconds post-stimulus, an effect that was not observed in the opposing “not-perceived” trials. In addition, we implemented a machine learning algorithm with a goal of using the pupillary data to predict whether conscious perception occurred on a trial-by-trial basis, which has reached about 71% accuracy, and continues to improve as we refine our results. In conclusion, our findings suggest that pupil diameter changes may be a good candidate for a physiological marker of conscious perception, which could be used both to further isolate the neural correlates of consciousness and to help assess the state of patients who are unresponsive, but possibly conscious.

Need vs. Supply Analysis of the New Haven Public Bus System

Stan Mathis, Yale University


The aim of the study was to assess how well the public transportation system of New Haven County was matched to the public transit need using publicly available geospatial datasets from state and federal sources. Geospatial bus stop data was extracted from public State of Connecticut data sets. Census tract geography was extracted from US Census TIGER files while census tract aggregated household vehicle access data was queried from the American Community Survey data access server. A census tract’s Need was defined as percentage of households reporting access to zero vehicles; its supply was defined as the number of bus stops per square mile of census tract. Analysis found pockets of need in six communities across the county. Need was mapped to supply by dividing tracts into need above and below the median and supply above and below the median. This defined areas of Over Served (low need, high supply), Good Fit (high need, high supply), and Under Served (high need, low supply). This stratification was mapped, exposing strongly underserved pockets in five communities and one anomalous neighborhood.

Penetrance estimates for incidental genomic findings in ACMG-59

James A. Diao, Yale University

The dropping costs and rising popularity of next-generation sequencing has introduced the possibility of personalizing medical treatments and screening for genetic diseases. Still, the clinical community’s understanding remains incomplete, with limited consensus on the proper interpretation for many genetic variants. Thus, the standard procedure when returning sequencing results has been to report findings only in genes related to the diagnostic indication, and not incidental findings in other genes. To balance the threat of false positives with the medical benefits of true findings, the American College on Medical Genetics and Genomics (ACMG) recommends an exception: that clinical sequencing laboratories seek and report incidental findings in a specific set of 56 genes (ACMG-56), in which variants are considered to have a greater likelihood of causing disease.

The clinical value of these recommendations are evaluated using the metric of penetrance, defined as the probability that a patient who tests positive will later develop the disease. We queried the public 1000 Genomes Project to obtain whole genome data for 2,504 healthy individuals from diverse ethnic populations. Pathogenic variants were identified using the central repository ClinVar, and found to be distributed unevenly across ancestral groups in this cohort, and incidental findings were found to be inflated relative to empirical disease prevalence. Quantitative risk estimates were derived by modeling penetrance as a function of disease prevalence, allele frequency, and allelic heterogeneity. Plausible ranges for these parameters were estimated from the 1000 Genomes Project cohort and the medical literature. Under the most generous assumptions, the maximum overall penetrance estimates for the majority of diseases fall under 50%, with many under 5%. Penetrance estimates were also shown to vary significantly between ancestral groups, stemming from allele frequency differences between these groups.

Safer Chemicals Design Diagrams

Longzhu Shen, Yale University
Fjodor Melnikov, Yale University
John Roethle, Yale University
Aditya Gudibanda, Yale University
Richard Judson, US EPA
Julie Zimmerman, Yale University
Paul Anastas

The NRF2-ARE antioxidant pathway is an important biological sensing and regulating system that responds to chemical insults. At minute level, it protects a living species to go through hard environmental conditions. However, when the external disruption exceeds the inherent resilience, cellular damage can occur, eventually leading to cytotoxicity. Therefore, studying the likelihood of a chemical activating the NRF2-ARE pathway is interesting to discovering therapeutic agents and designing safer chemicals. In this research, we engaged a combination of computational chemistry, statistical learning and mechanistic toxicology to estimate the likelihood of a chemical to perturb this critical toxicological pathway and derive a scheme to guide chemical design with safer potency.

The Effect of Virtual Civic Engagement on Crime: SeeClickFix in New Haven

Daniel Spakowicz 4328949, Yale University
Carolyn J. Presley, Yale University
Dowin Boatright, Yale University
Mark Gerstein, Yale University
Ann Greene, Yale University
Marjorie Rosenthal, Yale University
Andrew V. Papachristos, Yale University

Mobile virtual communities are an emerging space for improving social cohesiveness and promoting collective efficacy. The application SeeClickFix is a smartphone and web application developed in New Haven, Connecticut, where users report issues in their communities including non-violent crimes. These posts can be supported and commented on by other users and local government agencies acknowledge and address issues. The data are publicly available, providing a data-rich and transparent venue for monitoring the interaction of individuals with each other and city representatives. The purpose of our study is to look for correlations between SeeClickFix use and crime. We hypothesize that SeeClickFix activity reduces crime by increasing social cohesion and promoting collective efficacy. Preliminary analyses show that within each neighborhood, months with more SeeClickFix posts tend to have fewer crimes. In addition, the crime rate is lower after the creation of SeeClickFix relative to before. These data suggest that SeeClickFix use is correlated with reduced crime in New Haven. Further efforts are needed to establish if there is a causal relationship and if so by what mechanism. This work has the potential to suggest a method by which communities can increase transparency and reduce crime through an open data platform.

Urban Environmental Performance Index: The Quito Pilot Case

Diego C. Manya, Yale University
Ryan Thomas, Yale University
Matthew Moroney, Yale University

In the context of the adoption of the Sustainable Development Goals (SDG) and its indicators, especially Goal 11, the importance of indicators is paramount to the implementation, monitoring and eventual success of the SDG. However, a review of global sustainability indicators reveals major gaps in coverage, including scientific gaps. Indicators are often siloed, meaning at the urban scale where issues often overlap and intersect, existing indicators do not take into consideration cross-sectoral linkages. Finally, many indicators fail to adequately address social equity or inclusion, a high-priority issue identified as central to new definitions of green economy and sustainable development at Rio+20. This research we developed a case study to pilot new indicators based information available for the city of Quito that address these challenges.

The development of this pilot indicators has involved the use of several data collection tools and particularly the treatment of data so it could be available for numerical and graphical analysis. The research relied heavily Open Source platforms and tools to perform the analysis. Ultimately, the objective of this research is to scale the study to other cities around the world and other indicators to create an index that is reliable in measuring the environmental performance of the cities, while considering the social distribution of the environmental impacts inside the city.

Yale’s Environmental Performance Index: the construction and use of a composite index for global sustainability

Zach Wendling, FES