Interdisciplinary – Page 45 – UROP Spring Symposium 2021

Interdisciplinary

Replication of Genetic Influence on Risk Aversion

Humans vary a lot in their willingness to take risks. An individual’s willingness to gamble, use recreational drugs, or engage in risky financial behavior, among other behaviors, are a part of a characteristic that researchers and academics call risk aversion (Linnér et all, 2019). Researchers have begun to understand the contribution of an individual’s genetics in explaining varying levels of risk aversion. In our study we try to replicate and build on the body of research regarding risk aversion and genetics. Using data from the Health and Retirement Study (HRS), a longitudinal study following American elders and retirees from 1992 to the present, over 20,000 individuals are asked to hypothetically gamble their future income, in exchange for increasing their lifetime income. The income gambling questions allows us to observe how much risk each individual is willing to take. We wanted to replicate the finding of a large 2019 study identifying 124 specific single-nucleotide polymorphisms (SNPs; Linnér et all, 2019). We used polygenic risk scores (PRS) to determine how genetically prone each individual is to risk aversion. We then checked if these individuals displayed risk aversion in the HRS data. By replicating the work of other researchers, academics will have a better idea of how individuals engage in risky decision making and a better understanding of the biological underpinnings of risk aversion. The study will also help researchers in the genoecomics field build credibility.

Replication of Genetic Influence on Risk Aversion Read More »

Coronary artery segmentation for automatic stenosis detection

One type of coronary artery disease (CAD), stenosis, is characterized by the narrowing of the coronary arteries, which is the leading cause of death in the United States. One of the most popular ways to diagnose stenosis is the coronary angiogram, from which a doctor can diagnose stenosis by finding places of narrowing in the arteries with the video or the pictures acquired. A lot of studies are currently focusing on automatic coronary artery segmentation, a critical step of a computer-aided system that assists doctors in detecting coronary stenosis. Here we propose a deep learning pipeline using DenseNet-backbone U-Net for coronary artery segmentation in angiogram images, which could be combined with pre-processing and post-processing steps for stenosis detection.

Coronary artery segmentation for automatic stenosis detection Read More »

Consentful Messaging

In the offline world, people’s communication with others revolves around networks. People tend to communicate more with people they have a strong tie with, which is easy to accomplish in fluid and nuanced ways. However, current social media systems lack such mechanisms for controlling interaction and communication based on network strength, which often leads to massive online harassment and abuse. A major example is Twitter, a social platform well-known for its openness. In this work, we present a system called NetRule, a Chrome extension that augments Twitter and gives users the ability to author network rules to control incoming messages and notifications. Through NetRule, users can easily combine and apply network-based rules on Twitter accounts that initiate interactions, such as whether the number of mutuals is high enough, whether the account has been blocked by one’s following, etc. Users also have the freedom to decide what happens to the accounts flagged by the rules, such as muting, blocking, or visibly coloring the accounts on the interface as a warning. Our evaluation of the system, a field deployment study on Twitter, showed that XXX.

Consentful Messaging Read More »

COVID-19 bioinformatics research

Bioinformatics has been a powerful method to study COVID-19 and our project look at ontology-based application in rational drug design. In our previous study (https://www.nature.com/articles/s41597-021-00799-w), the Coronavirus Infectious Disease Ontology (CIDO) was used as an ontological platform to represent anti-coronaviral drugs, drug targets, host-coronavirus interaction (HCI), and their relations. A “HCI checkpoint cocktail” strategy was further proposed to interrupt the important checkpoints in the dynamic HCI network and ontologies support this design process. However, the users such as drug researchers might not have the required ontology knowledge to use the information represented by the CIDO. Therefore, our project aims to design and build a user-friendly tool/website to allow basic queries and facilitate rational drug design using “checkpoint cocktail” strategy. We have developed a MySQL relational database that systematically represent the drugs, bioentities, interactions, pathways, and their relations. A set of real life data were added to the database. MySQL queries were performed to demonstrate our capabiliities to query different contents from the database to support the query of anti-coronaviral drug information. A web interface is being developed in order to further support online query and analysis of anti-coronaviral drugs, leading to rational anticoronaviral drug cocktail design.

COVID-19 bioinformatics research Read More »

Sentiment Analysis of International Trade Agreements

The aim was to examine the changes in sentiment that occurred within trade agreements over time using rule-based sentiment analysis. In this case, the sentiment of a text was measured by the degree to which it expresses or implies an opinion. The focus was on a collection of English hundred trade agreements written within the last few decades. Before analysis, a dictionary of trade terms was created using seminal texts. Terms were categorized based on if they expressed a cooperative, punitive, or bureaucratic sentiment. The analysis entailed assigning sentiment scores to trade agreements based on the number of categorized terms within. There is no clear expected result, but there will likely be some observable change in overall sentiment over time, whether it is more or less sentiment.

Sentiment Analysis of International Trade Agreements Read More »

What is on the table? Peer effects within A Large Canadian Restaurant Chain

Jianglai Zhang Pronouns: she/her/hers Research Mentor(s): Luchi He, Research Assistant of Dr. Yue Maggie Zhou Research Mentor School/College/Department: Strategy department, Ross School of Business Presentation Date: Thursday, April 22, 2021 Session: Session 1 (10am-10:50am) Breakout Room: Room 20 Presenter: 4 Event Link Abstract For privacy concerns this abstract cannot be published at this time. Authors:

What is on the table? Peer effects within A Large Canadian Restaurant Chain Read More »

Effective ways of communicating research on language learning by children and adults

Media use has been a beneficial tool for education in schools, especially over the past year, and can expand the knowledge that can be known by any individual person. While it advances education in the school systems, it may also have some negative effects regarding how children adapt to language acquisition, especially those who are not native to English in the U.S. In this study, we observed 23 children from Spanish-speaking homes. We asked about both the children’s and their parents Spanish and English media use through a questionnaire asking them to report their time spent on certain categories of media. In addition, two animated videos aiming to teach English possessive nouns and verbs were shown to the students. They were tested on English vocabulary, through the Peabody Picture Vocabulary Test, Spanish vocabulary through the Spanish version, TVIP, and on English and Spanish grammar through morphosyntax tests. The results of this project reinforce that educational content is beneficial for children’s language development. However, the results also show that parent’s media use negatively impacted childrens English morphosyntax scores, as well as their English and Spanish vocabulary knowledge. For future research, it would be beneficial to understand which aspects of educational media benefit bilingual children’s ability to develop their non-native language. Additionally, it would be interesting to see the impacts of interactive educational media compared to passive animated media.

Effective ways of communicating research on language learning by children and adults Read More »

Effective ways of communicating research on language learning by children and adults

Language can be taught in a variety of ways, but is there a specific way that helps young children benefit the most? The present study of English language learners was conducted in order to see if children’s acquisition of possessive and past tense in English could be facilitated by watching “passive” but targeted child-centric animations, and if a child’s first language plays a role in their second language acquisition. For this study, native Spanish- and Mandarin-speaking children were tested on their English vocabulary and grammar before and after being shown immersive language-learning animations where both implicit and explicit teaching methods were used. Although this study is still in progress we found that both the Spanish and Mandarin-speaking children were able to learn English vocabulary and grammar from the animations. One question that we now have is whether children who received the same exposure to the target learning material but in an interactive game would be better able to learn and whether their learning would be affected by the amount of media exposure children get in their day to day lives.

Effective ways of communicating research on language learning by children and adults Read More »

Data Gathering Data Analysis Toward Better Air Quality Outcomes in SW Detroit

In support of the ongoing initiatives by community advocates to cut diesel emissions in Southwest Detroit, this research aims to gauge the effects on air quality that commercial vehicles have in residential areas when using their streets as routes. SW Detroit has some of the highest levels of PM and toxic pollutants measured in Detroit, with Detroit having the worst air quality conditions in Michigan. This study accesses the ambient air quality in residential areas that have a high frequency of commercial trucks traveling their streets. We want to know how each passing truck contributes to the air pollution on the street it is using.

Data Gathering Data Analysis Toward Better Air Quality Outcomes in SW Detroit Read More »

lsa logoum logo