ARISE Colloquium 2020 | NYU Tandon School of Engineering

ARISE Colloquium 2020

Celebrating the achievements of our ARISE 2020 class online

Welcome to the 2020 ARISE Colloquium! This year our 62 students will be presenting in four concurrent Panel live-streams, which you can RSVP for here. We encourage attendees to stay for the duration of the Panels and to ask questions at the conclusion of each presentation. 

students in lab coats

Panel A


Students in Panel A listed alphabetically: 

Session 1

  • Andy Huang
  • Jibran Khan
  • Vanessa Lin
  • Jessica McDonald
  • Yarileldy Payano
  • Emma Pereira
  • Nyah Smith
  • David Ugo-Omenukwa

Session 2

  • Tenzin Choezin
  • Subah Mahbub
  • Edwil Philippe
  • Raina Plaisir
  • Nolan Shaffer
  • Sehajpreet Singh
  • Uchechukwu Uwanaka

Yarileldy Payano, Emma Pereira

group 1 arise

  • Lab: Privacy and Security Automation Lab (PSAL) [Computer Science & Engineering]
  • Faculty: Professor Rachel Greenstadt
  • Mentor(s): Kejsi Take, Cameron Ballard


With the increased use of social media platforms, misinformation has made its way into almost all aspects of society. Twitter, being one of these platforms, allows users to center their tweets around popular issues through hashtags. We focused our analysis on the use of misleading hashtags that further the spread of disinformation. In this study, we chose two prevalent topics like American politics and climate change to describe how misinformation is spread. We gathered tweets associated with common search terms for each topic and selected a subset for manual analysis. Our research proposes a method of classifying tweets with specific hashtags to measure their influence in its intended community. By identifying the content that receives the most attention we hope to illustrate what kinds of false information are most impactful across different platforms and issues.


Andy Huang

Andy Huang ARISE

  • Lab:  Composite Materials and Mechanics Lab [Mechanical Engineering]
  • Faculty: Professor Nikhil Gupta 
  • Mentor(s): Gary Mac


The advancements in 3D printing technology have revolutionized the additive manufacturing industry and the process has been adopted across all aspects of the industry. The 3D printing process involves converting a Computer-Aided Design (CAD) file to a format readable by the printer and the printer will build the part layer by layer. Fused deposition modeling (FDM) printers is the most common method of 3D printing and it is widely used by both consumer-oriented product designers and hobbyists. The unique characteristic of being a cyber-physical system and layer by layer build method allows malicious actors to embed hidden defects into the design without detection. Along with hidden defects, other risks exist that may cause harm to the cyber-physical environment. To examine the potential risks, the mechanical testing of objects with controlled internal defects, such as depositing filler material instead of modeling material at the defect site, were used and printing orientation was altered: printed in the x-direction, z-direction and 45°-direction. The potential for lower strength and detection evasion was measured with finite element analysis and ultrasonic inspection. Additionally, cyber-security breaches also pose as another threat. Statistical methods are able to approximate security breach possibilities but recent HACK3D hackathons have proved that skilled hackers are able to launch innovative cyber and physical attacks. A series of HACK3D challenges were given to participants as a form of crowdsourcing, and were used to investigate the attacks and study the pathway in threat taxonomy. In the end, the results confirm that securing 3D-printing cyber-physical systems is an extremely difficult task.


Nyah Smith, David Ugo-Omenukwa

bilab arise

  • Lab: Future Building Informatics and Visualization Lab (biLAB) [Civil & Urban Engineering]
  • Faculty: Professor Semiha Ergan
  • Mentor(s): Zhengbo Zou, Daniel Bin Lu, Xinran Yu


People, on average, spend 90% of their time indoors. However, our static environments do not currently have the capabilities to change and adapt themselves in order to match our emotional needs. They are not “smart” enough to recognize and respond to our changing emotional stimuli, to soothe and comfort us in times of distress, possibly harming our mental health in the long-term. Despite the promising outcome from previous research efforts in obtaining physiological data from visual stimuli, there is still a fundamental gap between obtaining this physiological data and using it to predict the emotions of the user. Furthermore, there is also a lack of research in the area of using emotion detection and recognition of occupants to modify the indoor environment accordingly. This paper aims to rectify both of these gaps by first using Electroencephalography (EEG) signals to measure and infer emotions through machine learning models, and then use them to promptly modify architectural design features using adaptive VR environment setups.


Jessica McDonald

Jessica Mcdonald ARISE

  • Lab: MARL [Electrical Engineering]
  • Faculty: Professor Juan Bello
  • Mentor(s): Magdalena Fuentes, Charlie Mydlarz


The Sounds of New York City (SONYC) project has recorded more than 50 years' worth of audio data from around New York City. Microphones throughout the city have picked up audio in order to better understand the frequency, volume, and specific types of noise, and to apply that knowledge to relevant projects dealing with traffic, noise complaints, and more. Once they are recorded, the ten-second audio clips are annotated by volunteers through the Zooniverse platform, who label the sounds they hear. The audio data tells us a lot about the dynamics of the city, including when people are the loudest, traffic, and music played on the streets of the city. However, the music aspect of the SONYC dataset has not been previously explored. This research focuses on creating a curated dataset of music played throughout the city using the SONYC dataset. To do this, we sorted the audio using the volunteer annotations and filtered out irrelevant audio clips (i.e., ones with no music present), and analyzed temporal trends in the data. The goal of this project is to make a useful dataset that can be used for further research and/or machine learning surrounding music in New York City.


Jibran Khan, Vanessa Lin

jibran khan vanessa lin arise

  • Lab: Biointerfacial and Diagnostics Lab [Chemical and Biomolecular Engineering]
  • Faculty: Professor Rastislav Levicky
  • Mentor(s): Vladislav Frenkel


The electromagnetic spectrum consists of a range of light, including the kinds that aren’t visible to the human eye, such as radio waves, microwaves, infrared radiation, ultraviolet rays, and more. The small section of the electromagnetic spectrum that we can see is visible light. White light is a combination of all the colors of the visible light spectrum, with each color having a specific frequency and wavelength. Spectrometers are able to break the light from a single source into its component colors based on their wavelengths. The goal of our project is to build and iterate on a design of a simple spectrometer using household materials such as duct tape and black paper. A spectrometer has three major parts: the slit, the diffraction grating, and the viewing port. The slit allows light to enter the spectrometer, while the viewing port allows you to view the emission spectra. The diffraction grating is made up of angled grooves that produce the spectra based on the diffraction and interference of light. The household spectrometers will be used to analyze different light sources in order to understand how our eyes perceive them. We can then compare different light sources and intensities and see how these factors affect the spectrum that is produced. Furthermore, the spectrometer will be implemented into a larger range of applications, such as passing light through a solution to find the concentration by comparing the transmitted spectra.

Nolan Shaffer, Sehajpreet Singh

shaffer singh

  • Lab: Molecular and Cellular Biology Lab [Biology]
  • Faculty: Professor Fei Li
  • Mentor(s): Qianhua Dong, Hyoju Ban


Centromere is necessary for chromosome segregation during cell division. Kinetochore proteins are assembled on centromeric DNA through the fundamental histone H3 variant, CENP-A. The outer kinetochore proteins, such as Ndc80 and DASH/Dam1 complexes, interact directly with microtubules to segregate chromosomes. In our previous genome-wide visual screening, we found mutations of dad2, spc19, spc34 and dad5 which are the components of DASH/Dam1 complex that affect the distribution of CENP-A in fission yeast. However, we did not find all of the ten components of the DASH/Dam1 complex from the screening. This raises the questions whether the whole complex is required for CENP-A localization; why and how outer kinetochore proteins change the distribution of CENP-A. In this study, we will investigate the CENP-A-GFP distribution, centromere integrity in each of the DASH/Dam1 complex subunit mutants and some of the double or triple mutants. This study may help define the role of the DASH/Dam1 complex in the distribution of CENP-A. Mis-localization of CENP-A can cause severe chromosome segregation defects which results in aneuploidy or cell death. Understanding the mechanism of CENP-A distribution regulated by fungus-specific DASH/Dam1 complex might contribute to developing a potent antifungal drug.


Uchechukwu Uwanaka

uche arise

  • Lab: Machines in Motion Laboratory [Electrical Engineering]
  • Faculty: Professor Ludovic Righetti
  • Mentor(s): Sebastien Kleff, Avadesh Meduri


As several advancements in technology appear everyday, scientists and engineers are constantly looking for ways to employ these new systems in the most efficient way possible. The rise in autonomy for robots is one instance. With increased autonomy, robots can be used in transportation, power supply, and fault recognition/recovery. The emerging method of multi-agent systems (MAS), with the potential for maximum autonomy, became an option against the previous object-oriented programming. Many articles have delved into this idea, accurately describing the architecture, software, and comparisons that arose. However, some articles fail to provide a sufficient review of MAS without bias and conflicts of interest. This is seen with the article, “A decentralized multi-agent control approach for robust robot plan execution” by Li P, Yang S and Wang S.” that we evaluated in this poster. This article fails to provide a good comparison between MAS and object-oriented systems, by focusing less on the challenges of MAS and variations in its software. This oversight is attributed to a conflict of interest on the part of one of the researchers. In this poster through examining other works on the topic, we provide an effective review of MAS and a corrected comparison with object-oriented systems.


Subah Mahbub, Raina Plaisir

Environmental engineering arise

  • Lab: Environmental Engineering and Water Treatment Laboratory [Civil & Urban Engineering]
  • Faculty: Professor Andrea Silverman
  • Mentor(s): Mwale Chiyenge, Catherine Hoar


Combined sewer systems found in cities like New York collect and convey both stormwater runoff and sewage water in a single system. During high intensity rain events, these systems may become overburdened and backlogged, resulting in street-level flooding. This floodwater often contains sewage and its contaminants, including harmful fecal-related bacteria. To explain the impact of such flooding, we are conducting a literature review on the impacts of fecal pollution on urban surfaces with the goal of designing experiments to investigate the deposition and die-off of sewage-related bacteria on flood-impacted surfaces. This literature review is essential as flooding is on the rise due to climate change. With more frequent heavy rainfall events we expect to see increased discharge of sewage-containing water directly to the environment. The literature review will help us develop a standard methodology that our lab can use to study the impacts of bacteria deposited on urban surfaces after a flood on the microbiome of these surfaces. In conclusion, the ultimate goal of this review is to raise awareness of the potential risks regarding bacteria deposition on urban surfaces from combined sewer flood water and to motivate people to understand the potential health hazards related to this issue.


Edwil Philippe

Edwil Philippe arise

  • Lab: Dynamical Systems Laboratory [Mechanical Engineering]
  • Faculty: Professor Maurizio Porfiri
  • Mentor(s): Mert Karakaya, Roni Barak Ventura


Electronic waste (E-waste) represents more than 70% of the overall toxic waste in the United States. Even though a large amount of E-waste consists of reusable working electronic parts, E-waste is typically not recycled for the production of new technological devices. In this project, we salvage E-waste parts and build a robotic fish. The robot’s movement is controlled with an interactive phone application, connected via bluetooth. The robot will be used for educational purposes, demonstrating to the public the ubiquity of engineering and innovation, and robotics in particular.


Tenzin Choezin

Tenzin ARISE

  • Lab: Flow Chemistry with Microsystems Laboratory [Chemical and Biomolecular Engineering]
  • Faculty: Professor Ryan Hartman
  • Mentor(s): Tianyi Hua


Methane hydrate stores tremendous amount of energy which is estimated to be equivalent to twice of all the other fossil fuels combined. In the deep bottom of the ocean, where methane hydrate naturally exists, microbes and worms have established complicated nutritional symbioses. Microbes not only consume and produce methane as their energy source and metabolism waste, but also produce biosurfactants that can catalyze natural gas hydrate formation or dissociation. The stability of methane hydrate is crucial as the release of methane gas from the bottom of the ocean can cause global warming. The focus of this study is to figure out the influence of different biosurfactants on methane hydrate stability. We model the methane hydrate dissociation kinetics with the catalyzation effect offered by theses microbes to predict the phase boundary behavior of methane hydrate against temperature fluctuations.

Panel B


Students in Panel B listed alphabetically:

Session 1

  • Jay Chi
  • Youwen Duan
  • Melissa Fonseca
  • Jason Harris
  • Nour Kandil
  • Bintia Keita
  • Most Parvin
  • Saadat Rafin

Session 2

  • Sahithi Attada
  • Raisa Bhuiyan
  • Aysha Naveed
  • Rachel Rose
  • Tenzin Sherab
  • Taufiqa Tajnin
  • Tahsin Uddin
  • Emily Willet

Bintia Keita

binitia arise

  • Lab: Center for Advanced Technology in Telecommunications (CATT) [Electrical Engineering]
  • Faculty: Professor Shivendra Panwar,
  • Mentor(s): Fraida Fund, Ashutosh Srivastava


Internet users living under oppressive regimes rely on anonymity in order to speak freely on the internet without fear of persecution. Tor is a protocol that enables anonymous use of the internet, however, its high latency (delay) leads to a poor user experience. To explore this issue, I will collect data on Tor performance under challenging network conditions, and also evaluate whether existing solutions for reducing latency can help. This research could help make Tor more usable for Internet users who require anonymity.


Melissa Fonseca, Saadat Rafin

fonseca rafin arise

  • Lab: Applied Dynamics and Optimization Lab [Mechanical Engineering]
  • Faculty: Professor Joo H. Kim
  • Mentor(s): William Z. Peng, Hyunjong Song, Hyun Seok Shin


In order for a humanoid robot to navigate its environment safely, it must be able to stabilize itself or, if falling is unavoidable, prepare for a fall. Existing biped balance controllers often rely on center of mass (COM) state regulation for self-stabilization. While algorithms and sensors exist for real-time COM state estimation, COM-based fall prediction remains an active research area. Here we investigate the balanced region of a biped robot for standing push recovery in COM state space — the set of all initial COM states from which a biped can maintain its balance while remaining in a standing pose. Within the open-source Webots simulation environment, various initial COM states are imposed on the DARwIn-OP humanoid robot to determine the magnitudes of the velocity perturbations required for it to fall, which provide a numerical estimate of the balanced region. Both ankle and hip strategies were considered for push recovery control. Once computed offline, the balanced region obtained can be used within a real-time COM-state-based controller to enhance the controller’s balancing capability.


Most Parvin


  • Lab: The Chromosome Inheritance Lab [Biology]
  • Faculty: Professor Andreas Hockwagen
  • Mentor(s): Andreas Hochwagen, Carolyn Milano


Meiosis is a specialized type of cell division causing haploid gametes to be produced. Meiotic chromosome inheritance is very important because in addition to maintaining the chromosome number, meiosis also results in chromosome recombination creating genetic diversity. Failure of meiotic recombination could result in infertility or birth defects due to an abnormal number of chromosomes, such as Trisomy 18. To further understand meiotic recombination, I am investigating the role of a meiotic chromosome structure in chromosome inheritance using yeast, a single cell eukaryote. The mechanisms of meiosis are well conserved between yeast and humans. I am using tetrad analysis and genetics to analyze the role of one meiotic chromosome structure protein. Ultimately this work will give insight into the mechanisms of chromosome inheritance in humans.


Jay Chi, Nour Kandil

chi kandil arise

  • Lab: Biomolecular Engineering [Chemical and Biomolecular Engineering]
  • Faculty: Professor Jin Ryoun Kim
  • Mentor(s): Edward Chau, Emre Erkanli, Jineun Shin


Alzheimer’s disease (AD) has been linked to the misfolding and abnormal deposition of the amyloidogenic protein, Beta-Amyloid (Aβ), which is the primary component of extracellular brain plaques. Amyloidogenic proteins follow a typical pathway, in which soluble monomeric Aβ peptides interact to form toxic oligomers, and further aggregate to form insoluble fibrillar structures. This aggregation has been associated with the progression of AD pathology throughout the brain, leading to memory loss, disorientation, and mental decline. We investigated a select group of both natural and synthetic compounds that may act as inhibitors of Aβ aggregation by binding to and interacting with the protein. Using molecular docking techniques, we modeled how well our group of inhibitors bound to the Aβ peptides and its mutant variants, by comparing their binding energies and conformational compatibility. Implications from our study could contribute to uncovering the process behind amyloid aggregation, leading to discovery of new drugs for the detection and/or treatment of amyloidogenic diseases.


Youwen Duan

youwen duan arise

  • Lab: Dynamical Systems Laboratory [Mechanical Engineering]
  • Faculty: Professor Maurizio Porfiri
  • Mentor(s): Daniel Alberto Burbano Lombana, Jalil Hasanyan


Data-driven computational modeling is a fundamental tool in biology and engineering to characterize and predict animal behavior. This approach can be used for understanding how animals interact with their environment and how they communicate with each other. For instance, we can use mathematical models to describe fish navigation, which is a complex process that typically integrates multiple cues such as visual, olfactory, tactile, and hydromechanical. In this work, we will develop a physics-based computer application in Unity to reproduce the behavior of a fish swimming in a river. The main challenge is to obtain realistic fish motion while capturing interactions of the fish with the environment such as fluid flow, multiple obstacles, and interactions with other fish. To this end, we will leverage our recent mathematical findings in modeling fish behavior to simulate environmental and social interactions.


Jason Harris

jason harris arise

  • Lab: Urban Modeling Group Lab [Civil & Urban Engineering]
  • Faculty: 
  • Mentor(s): Debra Laefer, Frank Jiang


Essential Service Disparity during the NYC COVID-19 PAUSE order

COVID-19 quickly became a humanitarian crisis as the virus quickly bloomed into a global scale pandemic. The United States has been hit hard by COVID-19, especially New York City (NYC), as America’s first epic-center with as many as 11,571 cases in a single day. During this time, most low-income neighborhoods, already considered vulnerable according to the CDC’s social vulnerability index, faced especially devastating effects as these neighborhoods were disproportionately affected by COVID-19. This study aims to use the behavioral data of NYC near various hospital facilities in accordance with socio-economic status data and mobility reports among public transportation to determine how the paucity of essential services (e.g. banks, grocery stores, hardware stores) resulted in much higher density visiting patterns compared to communities where residents had a broader range of choice of destination. For this, cellphone-based footfall data will be used in conjunction with the mobility patterns recorded by Prof. Laefer’s National Science Foundation funded team in the vicinity of 5 hospitals across NYC.

Sahithi Attada, Tenzin Sherab

developmental genomics arise

  • Lab: Developmental Genomics Lab [Biology]
  • Faculty: Professor Christine Rushlow
  • Mentor(s): Peter Whitney, Chris Rushlow


Analysis of High Resolution Images of Gene Expression

Feature extraction for image analysis encompasses a broad set of computational techniques to interpret and quantify images. Continuous improvement in these techniques has allowed researchers in many fields of study to greatly expand experiments that normally would be limited by time-intensive manual annotation. However, many algorithms used in feature extraction produce some percentage of errors which limit the precision of some analyses. Here, we employ a number of different strategies to solve a feature extraction problem relevant to biological research. This problem involves assigning a feature associated with gene activation to a specific nucleus in a monolayer of cells imaged with confocal microscopy. We compare a Euclidean distance method, an image segmentation method, and hand-curation method, ranking these approaches by both accuracy and time-intensiveness. We will use this data to create a hybrid approach, one which uses manual intervention for ambiguous cases, but automatically annotates in cases of high confidence. This final approach will be compared to all three previous approaches as a potential strategy for a broad class of problems where 100% accuracy is required, and maximum time efficiency is desired.


Tahsin Uddin, Emily Willet

Uddin willet arise

  • Lab: Primate Hormones and Behavior Lab [Anthropology]
  • Faculty: Professor James Higham
  • Mentor(s): Christina Costa, Jessica Gunson


SARS-CoV-2, which originated in Wuhan, China, is a rapidly spreading virus that causes the deadly COVID-19 disease. The cell surface protein ACE2 (Angiotensin-converting enzyme 2) is the primary viral receptor of SARS-CoV-2, and different animal species display variation in their ACE2 sequence. Variation at key binding residues (amino acids) is thought to alter the structure of the ACE2 receptor and affect its interaction with the viral binding domain, increasing or decreasing the efficiency at which the SARS-CoV-2 spike protein can bind to and enter our cells. The interaction (binding affinity) between SARS-CoV-2 and ACE2 can be an important indicator of susceptibility to COVID-19: the more efficient the binding, the greater the likelihood of infection. This means that we can compare the protein sequences of our close relatives, such as catarrhines (monkeys and apes of Asia and Africa) and platyrrhines (monkeys of Central and South America) to predict their infection risk. A previous study found catarrhines share the same key binding residues as humans, suggesting they have a similar virus binding affinity as humans and may be equally susceptible. Platyrrhines, however, differ from catarrhines at 3 key sites which are predicted to greatly reduce binding affinity and susceptibility to the virus. These studies were conducted on a limited sample. To see if these statements hold true across a broader range of species, we will perform multiple sequence alignment on unpublished primate ACE2 sequences and examine variation at identified key binding residues to determine the relative susceptibility to SARS-CoV-2 infection of different catarrhine and platyrrhine species. Understanding which primates are at the greatest risk of infection will help us inform best practices for the conservation and management of these species in their natural environments.


Aysha Naveed, Taufiqa Tajnin

naveed tajnin arise

  • Lab: Soil Mechanics lab [Civil & Urban Engineering]
  • Faculty: Professor Magued Iskander
  • Mentor(s): Abdelaziz Ads, Linzhu li, and Sophia R Mercurio


The study of torpedo anchors moving through soil is of interest to geotechnical engineers because it allows for more information to be gathered, specifically soil deformation, assisting research and ensuring soil stability in infrastructure. This research aims to analyze the relationship between static and dynamic velocity of the torpedo penetration. Two shapes of torpedo nose were employed, conical and hemispherical. The analysis allows for better understanding and prediction of the dynamic response using static resistance. To conduct this research, digital image correlation (DIC) was used to analyze images of torpedo penetration. Images were taken at a frame rate of 2500 f/s. DIC compares consecutive pairs of images to each other and tracks the movement of embedded tracer particles. Shear-strain was calculated based on particle displacement. Results showed that the average dynamic resistance of fully embedded torpedoes is similar to the static resistance. Therefore, dynamic penetration resistance can be predicted from static tests. It was also concluded that in static events the nose shape of torpedoes does not have a significant effect on resistance. Thus, the critical parameter to predict dynamic resistance is the shear strength determined from static test results and independent of nose shape.


Raisa Bhuiyan, Rachel Rose

data responsibly lab arise

  • Lab: Data, Responsibly Lab [Computer Science and Engineering]
  • Faculty: Professor Julia Stoyanovich
  • Mentor(s): Kunal Relia


Computational social choice (COMSOC) is an interdisciplinary area of research that bridges computer science and social choice theory with an aim of understanding the computational aspects of aggregation of individual preferences toward a collective choice. Primarily, the research has focused on single-winner elections, where the aim is to determine one winner based on voters’ preferences. But, recently, there has been an increased focus on multi-winner elections, where the aim is to select a subset of candidates, also known as a committee, based on voters’ preferences. Specifically, a recent algorithmic framework discusses that a committee has three desirable properties: (i) diversity of candidates (ii) representation of voter populations and (iii) highest score. Diversity constraints are used to ensure appropriate representation of candidates being selected from each attribute group and representation constraints to ensure predefined voter populations are also represented in the committee. In our work, we aim to facilitate the implementation of this algorithmic framework to assess the effect of incorporating diversity and representation constraints in multi-winner elections. We will achieve this aim by developing first-of-its-kind real-world datasets, derived from voting in Eurovision song contest and on United Nations’ resolutions, where voters (e.g., countries in Eurovision contest) give preferences over candidates (e.g., performers in Eurovision contest), and candidates and voters each have at least one attribute (e.g., the genre of the song played by a performer and the geographical region a country belongs to, respectively). We eventually plan to make these datasets public to facilitate the COMSOC community.

Panel C


Students in Panel C listed alphabetically:

Session 1

  • Bariat Bashiru
  • Jillian Darcy
  • Maima Islam
  • Daisy Roberts
  • Lizbet Rodriguez
  • Josh Senior
  • Max Shalom

Session 2

  • Yewande Adesina
  • Anaya Bussey
  • Mahia Chowdhury
  • Ilona Lameka
  • Ruqaiya Mithani
  • Nina Piesanen
  • Eileen Ye

Bariat Bashiru, Maima Islam

student from the arise applied micro-bioengineering lab 2020

  • Lab: Applied Micro-Bioengineering Lab [Biomedical Engineering]
  • Faculty: Professor Weiqiang Chen
  • Mentor(s): Apratim Bajpai, Rui Li


Mechanical force between cells and microenvironment is increasingly believed to regulate the physiological functions in vascular systems which by nature are highly asymmetrical. However, the impact of intercellular and intracellular forces on vascular asymmetric formation is not completely understood. Here, using micropatterned endothelial cells on mechanical force sensing substrates, we demonstrated that mechanical forces within in-vitro microvascular tissue regulate asymmetric vascular endothelial cell arrangement. Tuning the cell-matrix and cell-cell forces using engineered topographic geometries and pharmacological treatments regulates vascular asymmetric morphogenesis in vitro. A study using mouse diabetic aortic endothelial cells in high extracellular glucose concentrations showed that diseased endothelial cells exhibited abnormal cell alignments, intercellular and intracellular forces. We further studied this phenomenon by modeling the vascular sheets as mechanical force-propelled active particles and confirmed that mechanical force drives asymmetric endothelial cell alignments in different tissue geometry. Together, these findings demonstrate that vascular asymmetric arrangement requires mechanical force induced by both intracellular contractility and intercellular force, and indicate that tissue geometry and mechanical force may serve as the inductive cues for vascular morphogenesis.


Jillian Darcy

jilian darcy arise

  • Lab: Behavioral Urban Informatics, Logistics, and Transport (BUILT) Lab [Civil & Urban Engineering]
  • Faculty: Professor Joseph Chow
  • Mentor(s): Srushti Rath


The open-source Multi-Agent Transport Simulation (MATSim) Virtual Testbed developed by C2SMART simulates an 8-million-person population covering the entire New York City (NYC). The virtual testbed that integrates various transportation modes including cars, trains, bus, bikeshare, taxi, and other for-hire vehicles that can be used by policy makers for planning and decision-making purposes. There is a growing need to evaluate urban last mile deliveries and use of cargo bikes in reducing air pollution and congestion. This project aims to gather data for freight demand that can be incorporated in the simulation testbed along with emerging mobility technologies. Using the existing freight flow data (obtained by the team from NYMTC’s Best Practice Model) the objective is to collect incoming commodity flows from various Freight Analysis Framework (FAF) zones to NYC for different years. We will analyze aggregate demand data to adjust freight flows between different origin-destination pairs in NYC as per requirements of the MATSim testbed. We will be using the Quantum Geographic Information System (QGIS) tool (free and open source) for spatial analysis, studying demand patterns and for visualization purposes.


Daisy Roberts

daisy roberts arise

  • Lab: Center for Advanced Technology in Telecommunications (CATT) [Electrical Engineering
  • Faculty: Prof. Shivendra Panwar
  • Mentor(s): Fraida Fund, Ashutosh Srivastava


COVID-19 has made home network performance especially critical, as people now rely more on Internet access to carry out their day-to-day activities. However, many people in the United States do not have access to high-quality networks, and especially to networks with low delay. A variety of techniques have been proposed by researchers to reduce network delay, which could potentially improve the quality of experience for Internet users who lack access to high-quality home broadband. One such protocol, called Copa, has shown promise in improving network performance for users with the "worst" connections, in preliminary tests by Facebook. I will build on those results, and investigate the extent to which Copa helps improve network performance for users who do not have access to high-quality home broadband, and whether it has potential to narrow the digital divide.


Max Shalom

student from the arise chunara lab 2020

  • Lab: Chunara Lab [Computer Science and Engineering]
  • Faculty: Professor Rumi Chunara
  • Mentor(s): Yuan Zhao, Vishwali Mhasawade


Populations in low socioeconomic communities and geographies usually have a higher burden of cardiovascular diseases (CVD). Using data from the national representative surveys we can disentangle the clustering effect of social determinants within individuals and their association with CVD outcomes. Additionally, we can pinpoint at which factors such as education level, poverty, and malnutrition, are most vital to improving upon urgently to have early prevention and reduce the burden of CVDs in the population. The aim, specifically, is to use data provided by the National Health Interview Survey (NHIS) to assess the association between individual and community-level social determinants and CVD, including heart disease and stroke, and use regression and machine learning methods to predict CVD using social determinants. To do this we will become familiar with Python coding, such as Pandas, NumPy and svm in scikit-learn packages to effectively make a summary of the data, such as getting the percentage of categorical variables (ethnicity, marital status), mean for continuous variables (age, income, etc.), plotting the data and make predictions.


Lizbet Rodriguez, Joshua A. Senior

rodriguez senior

  • Lab: Protein Engineering and Molecular Design Lab [Chemical & Biomolecular Engineering]
  • Faculty: Professor Jin Montclare
  • Mentor(s): Jacob Kronenberg


Organophosphates, also referred to as OPs, are toxic chemicals used as insecticides. These poisonous chemicals are utilized in both agriculture and military settings. Organophosphates pose substantially harmful effects to both nature and humans; OPs inhibit the activation of cholinesterase, causing acetylcholine to build up in the nerves, resulting in overactive interactions with fatal consequences. Phosphotriesterase (PTE) enzymes have been proven to function to mitigate the impacts of these OPs by operating as biocatalysts, breaking down present organophosphates. Using the computer software Rosetta, we aim to maximize the stability of PTE enzymes, making the protein structures more stable in order to have wider success in its industrial applications. Rosetta allows for the reengineering of PTE enzymes implementing a Monte Carlo Algorithm. This method creates random mutational alterations to PTE enzymes, generating more stable structures that deactivate organophosphates. We will then test out these new generated PTE structures in Rosetta and use PyMol, a protein-modeling interface, to visualize them.

Anaya Bussey

anaya bussey arise

  • Lab: Mechatronics, Controls, and Robotics Lab [Mechanical Engineering]
  • Faculty: Professor Vikram Kapila
  • Mentor(s): Hassam Khan Wazir


Cerebrospinal Fluid (CSF) is a clear, colorless, watery fluid that is found between the brain and spinal cord. CSF is used to help the brain with protection, nourishment, and waste removal. CSF has two different kinds such as, interleukin-6(IL-6) and interleukin-8(IL-8) which are found in all people. However, information regarding the geriatric population (ages 70 and up) and their levels of CSF IL-6 and IL-8 is sparse. In this retracted study, the researchers show an association between CSF IL-6 and IL-8 and depression in a geriatric population-based sample. Originally, the authors found that women with ongoing major or minor depression had lower levels of CSF IL-6 and IL-8 compared to women without depression. However, the paper was retracted upon the authors’ request because two spreadsheets with mismatching ID codes were merged by mistake and the conclusion was drawn based on the results of that data. The paper was republished after the researchers rectified the mistakes. The new conclusion was that women with ongoing major or minor depression had higher levels of CSF IL-6 and IL-8 compared to women without depression. Nine other women developed depression during follow-ups, but there was no correlation between their depression and their CSF IL-6 and IL-8 levels. The results demonstrate that if a female member of the geriatric population has higher levels of CSF IL-6 and IL-8 as compared to an average female member with no depression, there is a possibility that they may have ongoing minor or major depression. However, if the depression is recently developed, it will not automatically show in their CSF levels. This experiment is a starting point to analyze the CSF levels in members of the geriatric population and the researchers concluded that CSF IL-6 and IL-8 may play a role in depression later in life.


Ilona Lameka, Eileen Ye

search arise

  • Lab: Smart Energy Research (SEARCH) Group [Electrical Engineering]
  • Faculty: Professor Yury Dvorkin
  • Mentor(s): Yury Dvorkin, Jip Kim, Anwar Khan


Power outages in energy systems are unavoidable, despite the best effort of electric power utilities, and pose many risks to people living in cities. Those with certain medical or living conditions, especially heat-related problems like hypothermia, are affected by the inability to turn on an AC unit and will be hurt the most during a power outage. These electricity vulnerable people need their loss of electricity to be attended to first in the case of a widespread failure of electricity supply. To research and address this issue, we will use socio-demographic data to identify electricity vulnerable population groups and overlay these groups with Con-Edison’s power outage map in real-time. More specifically, we will identify the factors to determine electricity vulnerable people and find, refine, and use available data streams about New York City to categorize and rank various groups of vulnerabilities across all parts of the city. Using these factors, we will design and implement a ranking algorithm for repair crews to use when there is limited repair staff, as is the case during the COVID-19 pandemic. This ranking in combination with live power outage data in New York City will create a tool that can be used in real-time by Con-Edison and various city’s agencies to assist electricity vulnerable people across the city.


Nina Piesanen

nina piesanen arise

  • Lab: Visualization and Data Analytics [Computer Science and Engineering]
  • Faculty: Professor Enrico Bertini
  • Mentor(s): Daniel Kerrigan


One common machine learning task is to train a model to classify a data point. The model users may want to know why a model made the prediction that it did. That is, they want to be able to interpret the model or see explanations for its predictions. One particular kind of explanation is a counterfactual, which shows how a data point can be changed in order for the model to assign it a different label. Often with counterfactuals, we are interested in what the smallest change is in order to cause the model to change its prediction. In this project, we will develop a tool that lets users explore counterfactual explanations for a given dataset and ML model. We want this tool to allow the user to select an instance from their dataset, generate counterfactuals for that instance, and visually compare the instance with the counterfactuals.


Yewande Adesina, Ruqaiya Mithani

Hominin Skeletal Morphology Lab arise 2020

  • Lab: Hominin Skeletal Morphology Lab [Anthropology]
  • Faculty: Professor Scott Williams
  • Mentor(s): Jeffrey Spear


Understanding how the last common ancestor (LCA) of chimpanzees and humans moved, and thus the context from which humans’ unique bipedal locomotion evolved, is a major unanswered question in paleoanthropology. Humans’ closest relatives, chimpanzees, bonobos, and gorillas, all move using a combination of knuckle-walking (walking on all four limbs but bearing weight through the back of the middle section of the fingers rather than the finger tips or full hand), vertical climbing (such as shimmying up trunks of trees), and suspension (in which the animal hangs below branches rather than standing on a surface). But fossil evidence and behavioral observations of more distantly related primates have led many researchers to challenge the simplest explanation that the LCA moved similarly to humans’ closest living relatives, the African apes. Here, we quantify the morphology (shape) the distal radius (the part of the forearm that connects with the wrist) from digital scans of 29 living primate taxa as well as two fossil hominins closely related to humans using three dimensional methods. We explore how the shape of the distal radius differs among primates that use different forms of locomotion, and compare these different shapes to fossil hominins using phylogenetic comparative methods (statistical methods that account for the relationship among taxa). We will use ancestral state reconstruction methods to model the morphology of the LCA. We will test whether the morphology of the distal radius corresponds with different locomotor repertoires in living taxa by comparing Brownian Motion (BM; random evolution) and Ornstein-Uhlenbeck models (OU; evolutionary models based on the assumption that taxa evolve toward a number of adaptive peaks) of evolution. Building on the best-fit evolutionary model based on living taxa, we will test how the fossil taxa and reconstructed LCA fit into the identified adaptive peaks.


Mahia Chowdhury

mahia chowdhury arise

  • Lab: Urban Modeling Group [Civil & Urban Engineering]
  • Faculty: Professor Debra Laefer
  • Mentor(s): Debra Laefer, Frank Jiang


Destination as a Predictor of Touch Behavior during NYC’s COVID-19 PAUSE Order This research investigates the touch behavior of individuals exiting hospitals based on their final destination. Specifically, this study considers the number and type of items touched for those who went to restaurants versus those who went grocery stores. Both choice of destination and specific touch behaviors are considered over a nearly two-month period from Mar. 22 - May 19, 2020 at the peak of the COVID-19 outbreak in New York City. The bjects touched were split into five categories: “stationary environmental objects,” “personal objects,” “self,” and “shared objects.” Preliminary results indicate that people going to restaurants touched more objects than those going to grocery stores. This behavior was in part correlated to the time to the destination. The final analysis will consider a touch per minute score and a gender breakdown of the data.

Panel D


Students in Panel D listed alphabetically:

Session 1

  • Antara Afroz
  • Sydney Ho
  • Justin Hohn
  • Nancy Khosla
  • Sakeef Kibria
  • Natasha Piedrabuena
  • Sardorbek Rahmatulloev
  • Luisa Valdez

Session 2

  • Ayan Cooper
  • Diego Daza
  • Victor Ghosh
  • Tammy Li
  • Stephan McGlashan
  • Naomi Sergew
  • Edrick Torres
  • Valerie Viteri

Justin Hohn, Luisa Valdez

hohn valdez arise

  • Lab: Chromatin Genomics  [Biology]
  • Faculty: Professor Sevinc Ercan
  • Mentor(s): David Jimenez, Bhavana Ragipani, Sevinc Ercan


Long and linear eukaryotic DNA are packaged and organized inside the cell nucleus. This is done in part by DNA looping, a process mediated by the Structural Maintenance of Chromosomes (SMC) family of protein complexes. Condensins are SMC complexes that compact DNA in preparation for chromosome segregation during cell division. The Ercan lab aims to determine how condensins bind to DNA and function. To do this, the lab uses C. elegans, where a specialized condensin forms the dosage compensation complex (DCC). Dosage compensation is an evolutionarily conserved process that is accomplished in different ways across different species, however the outcome is the same in that it leads to similar levels of expression from X chromosomes between sexes. C. elegans DCC binds specifically to the X chromosomes in XX hermaphrodites to reduce their transcription two-fold, equalizing X expression to that of XO males. DCC is recruited to the X chromosomes at ~60 sites. Proteins that are required to recruit the DCC to these sites are also involved in repressing genes required for male development. We hypothesize that DCC recruitment sites may have evolved near male-specific genes located on the X. To test this hypothesis, we will use our existing data and publicly available databases and tools to determine if the ~60 DCC recruitment sites are found significantly closer to male-specific genes.


Sydney Ho

sydney ho arise

  • Lab: Visualization and Data Analytics [Computer Science and Engineering]
  • Faculty: Professor Enrico Bertini
  • Mentor(s): Jun Yuan


The analysis of text data remains a challenge for their intrinsic high dimensionality and the difficulty of providing context and semantics. With the help of visualization and machine learning models, analysts can understand a collection of documents more effectively and more efficiently. By conducting exploratory data analysis (EDA) on a set of documents, especially text grouping and comparison, analysts can generate a lot of insights to support further research on the text data. In this research, we focus on a collection of abstracts from IEEE Visualization publications and apply clustering algorithms to the text data. Then we construct a few visualizations to analyze how the publication content evolves over time, and how the research topics in different tracks are different from each other.


Antara Afroz, Natasha Piedrabuena

afroz piedra arise

  • Lab: Hybrid Nanomaterials Laboratory [Chemical and Biomolecular Engineering]
  • Faculty: Professor Ayaskanta Sahu
  • Mentor(s): Ingrid Joylyn Paredes, Steven Farrell, Haripriya Kannan


Sustainable energy is one of the most pressing issues that society faces today. With traditional materials reaching their performance limits, the development of new materials will lead us to meet the global demand for energy in a clean and sustainable fashion. Hybrid materials are an emerging class of compounds that combine inorganic and organic components to produce materials with performance greater than the sum of their parts. Our lab seeks to gain fundamental understanding of the electronic and thermal properties of these hybrid systems. We will do this via the synthesis and characterization of hybrid materials, looking at the effect of composition, size, and shape on thermal and electronic properties. Once reliable methodologies for synthesis have been developed, we will implement our materials into devices including light emitting diodes, solar cells, and thermoelectric coolers.


Nancy Khosla

Nancy khosla

  • Lab: Mechatronics, Controls, and Robotics Lab [Mechanical Engineering]
  • Faculty: Professor Vikram Kapila
  • Mentor(s): Hassam Khan Wazir


Walking is a very challenging task for robots and the problem of foot placement in particular, has been studied extensively. The retracted study considered for this project, proposed a novel control algorithm for finding the optimal step time and location for the robot to achieve a suitable walking gait from a viable location. In this study, the robot dynamics were estimated with a 3D linear inverted pendulum model and the constraint equations were analytically solved to determine the step time and step location of the robot. These equations were then simulated and based on the results obtained, it was concluded that it would take a finite number of steps for the proposed control algorithm to get to a desired gait cycle. Although the results obtained were very interesting and informative, the reason for the retraction of this study was that majority of the text and equations were plagiarized from a doctoral thesis and no credit or reference was given. Plagiarism is an unacceptable and unprincipled behavior because it is theft of someone else’s intellectual property for one’s own benefit. This paper, therefore, failed to meet the scientific standards set by the rest of the scientific community and therefore, it is very important that such activities are identified and reprimanded so that the integrity of the work done by other researchers can be kept intact.


Sakeef Kibria, Sardorbek Rahmatulloev

urbanMITS arise

  • Lab: UrbanMITS [Civil and Urban Engineering]
  • Faculty: Professor Kaan Ozbay
  • Mentor(s): Suzana Duran Bernardes, Jingqin Gao


The COVID-19 pandemic has disrupted transportation in New York City including: transit, private vehicles, ride-sharing, and bike-sharing. To better understand the effects of COVID-19 on a specific mode of transportation we focused on the bike-sharing system because of its growing popularity in urban cities around the world. During the early stages of COVID-19 bike-sharing in New York City had a peak in ridership. However, with the enforcement of social distancing and the stay-at-home order, the access to Citi Bike bikes had been limited, which has decreased bike usage and mobility around the city. To study this infrastructure, we hope to gather data from the Citi Bike database. We plan to use programming and data management tools, such as Python, SQL, and Pandas to depict bike usage trends prior to and throughout the pandemic. Additionally, we plan on identifying any fluctuations in bicycle “hotspots,” popular destinations, throughout the city. A possible method to locate hotspots is through the Citi Bike database and map oriented routes. We will also make use of extensive literature review to help us understand the main changes in behavior due to the pandemic. We hope this research will help contribute to our understanding of how the bike-sharing system functions and adapts to situations that impede transportation. Ultimately, our results will be used in the ongoing research on the impacts of COVID-19 on New York City’s transportation network performed by the UrbanMITS laboratory in conjunction with the C2SMART Center.

Ayan Cooper

ayan cooper arise

  • Lab: MARL [Electrical Engineering]
  • Faculty: Professor Juan Bello
  • Mentor(s): Charlie Mydlarz, Magdalena Fuentes


Noise pollution, which is widespread harmful noise levels, is a significant problem in urban areas across the United States. Estimates show that 9 out of 10 adults in New York City are exposed to immoderate noise levels, which extends across NYC’s 8.7 million residents. The Music And Audio Research Lab (MARL) and the Sounds Of New York City (SONYC)’s research is focused on tracking noise levels throughout NYC with the use of multiple sensors. Starting in March of 2017, SONYC’s goal is to identify what makes the city so loud, and how to use research to help those with noise issues understand their environment in terms of sound. In this project, we want to design a pipeline for the summarization and visualization of urban sound data in a way that helps residents of the city and city agencies better understand noise to enforce better NYC health policies. To do so, we examine hundreds of thousands of data points containing the measurement of Sound Pressure Level (SPL) and study its evolution over time at different temporal scales and locations.


Tammy Li, Stephan McGlashan

ece arise

  • Lab: ECE Machine Learnirng Lab [Computer & Electrical Engineering]
  • Faculty: Professor Anna Choromanska
  • Mentor(s): Apoorva Nandini Saridena


The field of self driving cars is being widely explored by researchers and companies, and much has been accomplished in this area already, of which Machine Learning is a huge part. We call our project “Training to Drive”. This project aims to understand the power of advanced Machine Learning algorithms such as Deep Learning, Supervised Learning by training a deep neural network to learn mapping from sensor data such as input images of road ahead of car to the car control commands such as steering wheel angle. We see rapid changes in the automobile industry especially in the last decade with the advancement of the Machine Learning field, new approaches are being developed to develop the driverless-cars to fix problems in existing approaches, in Training to Drive project, we hope to study one such approach that is popular widely i.e., end-to-end Deep Learning and Supervised Learning class of approaches and understand their limitations. The study is going to be empirical, so we collect data using an open-source driving simulator called Carla, which is a great software that allows us to collect driving data in realistic looking driving environments such as cities, different weather conditions, different road types, etc., and allow us to validate the developed models by deploying it in real-time simulation where we control the navigation of autonomous driving cars.


Naomi Sergew

naomi sergew arise

  • Lab: Behavioral Urban Informatics, Logistics, and Transport (BUILT) Laboratory [Civil & Urban Engineering]
  • Faculty: Professor Joseph Chow
  • Mentor(s): Srushti Rath


On-demand microtransit services have the potential to complement transit services allowing cities to provide flexible and convenient transportation options. Such services can improve accessibility to stations and destinations while helping people access jobs, education, healthcare, and social activities. The performance of various microtransit operations in different cities may be influenced by the city characteristics. The project seeks to gather socio-demographic, economic, housing, commute, and employment data (on census tract level) from American Community Survey (U.S Census Bureau) databases across different cities. In addition, the city boundaries, street network and transit network geo-spatial data will be collected. We will be using the free and open source Quantum Geographic Information System (QGIS) software to visualize aggregate information. The combination of all the city-level attributes collected for multiple cities can be used to build appropriate models for studying the demand and analyzing performance of different microtransit operations in a city.


Valerie Viteri

valerie viteri arise

  • Lab: Multifunctional Material Systems Lab [Chemical and Biomolecular Engineering]
  • Faculty: Prof. Miguel Modestino
  • Mentor(s): Adlai Katzenberg


The retraction of a scientific publication, Solar Photothermochemical Alkane Reverse Combustion by Chanmanee et al., was analyzed to learn about the importance of baseline experimental design and controlling variables properly. In the study, the authors wanted to gain a better understanding of the chemical processes to turn pollutants into fuel as a new renewable source. However, the scientists later discovered that impurities had affected the entirety of their data and led to its retraction. This analysis aims to comprehend the issues within the study, provide solutions to make the study viable, and compare it to accurate data. Through this investigation we can learn from previous mistakes to further our knowledge in the Chemical and Biomolecular Engineering field.


Edrick Torres

machines in motion arise

  • Lab: Machines in Motion [Electrical Engineering]
  • Faculty: Professor Ludovic Righetti
  • Mentor(s): Sebastien Kleff, Avadesh Meduri


In today's world technology has gone through many advancements and scientists all over the world spend night and day trying to make the world an easier and better place to live. With the rising advancements of autonomous robots, the dream of a better world is slowly becoming a reality. But some scientists in this community have abused the system to push studies that are not ready yet. This is seen in the Peer review scandal of the “Journal of Vibration and Control”. In this Scandal, 60 scientific research papers were retracted because of a flawed peer review system where the authors of these studies had used fake identities to review their work and have them published without going through actual peer review.  This was a major blow to members of the scientific community and demonstrated that only a few flawed individuals can give the scientific community a bad reputation. The 60 retracted papers show a big regression in the field of robotics and had the publication be legitimate, The scientific community and the world would be that much closer to a better world. This project will show the flaws in what these authors did and will demonstrate the correct and detailed way in which peer review should take place.


Diego Daza, Victor Ghosh

daza ghosh arise

  • Lab: Transformative Materials and Devices Laboratory [Mechanical Engineering]
  • Faculty: Professor Andre Taylor
  • Mentor(s): Jason Lipton, Jason Alexander Rohr


On October 2012, an article titled “Efficient Bulk Heterojunction Solar Cells using Tetrasubstituted Pyrene Derivatives as Donors” was published in the journal, Organic Electronics [Thomas, et al., 13 (10), 2201-2209, 2012]. The researchers behind the study claimed to have found an improved procedure for the fabrication of solution-processed bulk heterojunction solar cells, by using tetrasubstituted pyrene derivatives as electron donors and a solubilised version of the buckminsterfullerene, C60 as electron acceptor. They affirmed that when using this solar cell a Power Conversion Efficiency (PCE) of 3.63% could be achieved. While likely scientifically correct, the scientific community advocated for its retraction due to alarming concerns of plagiarism. The authors violated the professional ethical codes of the scientific community not only by making false claims of authorship of content already published in an article by the same journal [Dutta, et al., 13 (2), 273-282, 2012], but also by evading citing material taken from several different sources. In order to ensure that no papers are plagiarized, authors should be actively involved in the editing and reviewing of the articles, and Journals themselves should have stricter review and evaluation processes. By either executing the methods used in the original article or correctly replicating and citing the experiments previously performed, the authors could have reached the same results while still being able to follow the ethical principles laid out by the scientific community.