Workshop & conference agenda – Tuesday, Feb. 18, 2025
Presented by Theresa McKim, Ph.D., teaching assistant professor, biology, 69ÀÏʪ»ú¸£Àû in collaboration with the Nevada Bioinformatics Center
Python is a high-level, general-purpose, and open-source programming language. It is emerging as one of the most popular programming languages, both for scientific computing and general use. In this Python crash course, we will cover an overview of Python and programming fundamentals and then describe how to use Python for data processing, analysis and visualization. At the end of this module, you will be able to:
- Understand the basic components of a Jupyter notebook
- Write and debug basic Python code
- Use functions, loops, conditional statements, and apply programming fundamentals applicable across computing languages
- Analyze a dataset using Python libraries built for data science
- Learn and discuss best practices for data visualization to communicate your results
Presented by Hans Vasquez-Gross, Ph.D., Bioinformatician, Nevada Bioinformatics Center
This hands-on workshop is designed to help participants master the basics of the command line and UNIX fundamentals. Through interactive sessions, you’ll learn essential skills to navigate, manage files, and perform basic program installations.
- Get started with the command line interface
- Understand and explain the directory structure of computers
- Navigate across and within files and directories
- Create, copy and delete files efficiently
- Install programs using Conda
Whether you're new to the command line, looking to build your computational skills, or hoping to meet like-minded individuals, this workshop is an excellent place to start!
Presented by Keynote Speaker Martin Krzywinski
This workshop will distill the core concepts of information design into practical guidelines for creating visual explanations of science: figures, posters and graphical abstracts. The focus will be on clarity and concision and on the idea that form follows function. You'll learn visual strategies for organization, emphasis and theme. To illustrate these guidelines concretely, Martin will walk you through redesigns of scientific visualizations from the wild.
Data speaks for itself, but what is it saying? Crafting visual explanations of complex ideas.
Moderators
Katia Albright, director, Nevada Career Studio, University of Nevada Reno
Panelists
Education: B.S. Economics & Finance
Current Position: Cybersecurity Data Science Manager, Sierra Nevada Corporation; WiDS Ambassador
Tatiana is a data professional with 10 years of experience in analytics, specializing in forecasting, budgeting, research and compliance. She has a proven track record of productivity and quality in fast-paced environments. With strong analytical skills and attention to detail, she excels in troubleshooting data and technical issues, generating reports, training individuals of all levels, and optimizing business processes for efficiency. Tatiana is experienced in leadership, teamwork, and communication, making her a valuable asset in data-driven decision-making.
Education: B.S. Mathematics and Statistics; M.S. Statistics and Data Science
Current Position: Software Engineer, Light & Wonder
Alex is a data scientist with a passion for programming. A 69ÀÏʪ»ú¸£Àû alumnus, he earned his M.S. in Statistics and Data Science and has been engaged in bioinformatics and data science since joining the Nevada Bioinformatics Center in 2020. With expertise in software engineering and data-driven problem-solving, Alex continues to contribute to innovative projects in analytics and computation.
Education: A.S. Mathematics; B.S. Mathematics
Current Position: Statistician and Data Scientist, Northern Nevada Public Health
Lissa is a data analyst and statistician specializing in respiratory case data analysis at Northern Nevada Public Health. She develops R-based tools that require no coding experience, making data insights more accessible. Previously, she conducted energy research utilizing regression models. A lifelong Renoite and 69ÀÏʪ»ú¸£Àû graduate, Lissa is passionate about data science, analytics, and public health.
Education: A.S. Natural Sciences; B.S. Biology; M.S. Bioinformatics (Computer Science + Big Analytics)
Current Position: VP of Data and AI Governance, Upland Specialty Insurance
Jackie is a strategic leader in data and AI governance, leveraging technology and innovation to drive growth in highly competitive markets. She specializes in identifying and implementing cutting-edge AI, engineering and IT advancements to revolutionize workflows and expand business opportunities. Jackie has a diverse background spanning bioinformatics, wildlife biology and data science, with extensive experience in the insurance sector. She is also the organizer of the Big Data and Brews meetup, a key networking hub for local data scientists.
Education: B.S. Biology (Anatomy and Physiology); B.A. Anthropology; Ph.D. Anthropology
Current Position: Senior User Experience Researcher, Google DeepMind
Chris is a technology anthropologist with expertise in applied research across retail, healthcare, indigenous rights, substance use, e-commerce, ecological design, machine learning, and information technology. Currently at Google DeepMind, he applies anthropological insights to UX research, helping shape user-centered AI innovations. He holds a Ph.D. in Anthropology from the University of Pennsylvania.
Education: A.A. Business Administration and Management, B.B.A. Management Information Systems
Current Position: Field Chief Technology Officer, TrueMark
Derek has over 10 years of experience leading enterprise delivery teams, specializing in enterprise architecture, CI/CD pipeline orchestration, and DevOps. Previously, he was a Senior Solutions Architect at AWS, focusing on code quality and delivery velocity. His background spans mobile technology, aviation engineering, logistics, and operations. A former Special Forces Medical Sergeant, small business owner and sales associate, Derek brings diverse expertise in leadership, technology, and collaboration, with a strong commitment to quality and innovation.
Education: B.S. Mathematics and Statistics; M.S. Statistics and Data Science
Current Position: AI Engineer, IBM
Sonali is an AI Engineer with IBM’s Client Engineering team, specializing in developing data-driven software solutions across diverse industries. She holds a degree in Engineering Science from the University of Toronto, where she majored in Machine Intelligence, and is currently pursuing a master’s in data science from the University of Texas at Austin. A strong programmer, communicator and critical thinker, Sonali is passionate about advancing machine learning and looks forward to sharing her insights on the panel.
Presenters
Michael Martin, computer science undergraduate student, College of Engineering, 69ÀÏʪ»ú¸£Àû
Harnessing Machine Learning and NASA Satellite Big Data for Enhanced Wildfire Prediction and Air Quality Forecasting
Advancements in Machine Learning (ML) have transformed time series prediction, particularly in projection of extreme weather phenomena such as wildfires. This research aims to improve wildfire prediction using NASA satellite remotely sensed big data. We analyzed 25 years of NASA high-resolution environmental variables over the Western U.S. to develop ML-based models for enhancing accuracy of fire-induced air pollutants. We developed deep learning architectures to enhance the precision of wildfire-related emission predictions. Feature importance analysis using SHAP improves model interpretability, ensuring transparency in AI-driven forecasts. This approach is crucial for developing efficient, explainable data-driven models to support informed decision-making in environmental scenarios.
Arielle Pastore, anthropology graduate student, College of Liberal Arts, 69ÀÏʪ»ú¸£Àû
Re-thinking Ordinal Data in Dental Anthropology: Insights from Multiple Correspondence Analysis
This study re-evaluates the utilization of ordinal data in dental anthropology by employing multiple correspondence analysis (MCA) across four samples, including the Subadult Virtual Anthropology Database, Texas State University Donated Skeletal Collection, University of Tennessee, Knoxville Donated Skeletal Collection, and a European sample from Christy G. Turner. Eigenvalue analysis across these datasets reveals that the first principal dimension consistently accounts for 11.4% to 12.7% of total variance, indicating a recurrent pattern of dimensional importance. Explorations of inter-trait relationships among diverse datasets expose their critical role in morphological differentiation and highlight their utility as robust markers in population-level analyses.
Eriko Sakamura, M.S., M.A., instructor, Computer Science and Technology, Truckee Meadows Community College
Undergraduate projects at TMCC: What community college students can do with data science?
At TMCC, we have been active in data science-related undergraduate research, providing students with opportunities to apply analytical skills to real-world problems. Our projects span various disciplines, including biology, public health, and environmental science, allowing students to work with real datasets and develop critical thinking skills. In this talk, I will highlight ongoing undergraduate projects, including an INBRE-funded study on biodiversity and climate, as well as coursework-driven projects from DATA101 and DATA220. By showcasing these initiatives, I aim to demonstrate the potential of community college students in data science and how these experiences prepare them for future academic and professional success.
Timothy Malacarne, Ph.D., assistant professor of Data Science, Nevada State University
Tipping Hats, Riding Coattails: Reference Networks in Country Music
This project examines reference networks in American country music lyrics. References to icons, heroes, and ancestors provide rich material for narrative construction of self and one’s imagined community. This project uses a dataset of 74,000 American country music songs from the past 60 years to investigate artists’ use of lyrical citation of peers and predecessors, as part of a complex process of meaning-making at the individual and cultural levels. It has also served as a teaching resource for my courses in Nevada State University's Data Science program.
Drinks and appetizers will be served. Please visit our sponsor booths!
Preliminary Geothermal Play Fairway Workflow for the Great Basin Region, USA
Authors: Nicole Wagoner, Mark Coolbaugh, James Faulds, Elijah Mlawsky, Cary Lindsey, Whitney Trainor-Guitton
Play fairway analysis (PFA) is an exploration tool to assess geothermal resource potential and reduce geothermal resource exploration risk. Geothermal exploration risk is particularly high when searching for hidden geothermal systems (i.e., systems without surface expressions such as hot springs). Many hidden systems exist in the Great Basin region (GBR) of the western United States, a world-class geothermal province with over 1 GWe installed nameplate capacity. The INGENIOUS project aims to build on previous PFAs, as well as recent machine learning-based work to improve methodologies for discovering new, economically viable, hidden systems in the GBR. Here, we present a preliminary GBR play fairway workflow built from the assessment of 14 newly updated regional geological, geophysical, and geochemical datasets. The datasets have been analyzed with weights of evidence, logistic regression, and other tools to identify statistically significant relationships between data layers and known geothermal systems. Additionally, feature engineering has been utilized to extract maximum value from the data by developing hybrid predictive features consistent with previously identified physiographic relationships. The identified key predictive feature layers were then statistically integrated using PFA architecture into a preliminary GBR play fairway model. The results improve our understanding of GBR geothermal resources and facilitate identification of new hidden geothermal systems.
Examining Test Anxiety in Math Using R: Mean Profile Plot in Repeated-Measures ANOVA with Bonferroni Adjustment of Matched Sample T-Tests
Authors: Carson Perkins, Colleen I. Murray
We use R to demonstrate a repeated-measures ANOVA with a mean profile plot and then conduct a post-hoc comparison of means using matched sample t-tests with Bonferroni adjustments. Analyses are based on a sample of heart rates of students at baseline, before a math test and after a math test.
The Practicality of Japanese Tests Created in One Hour Using AI: Teaching Experience or AI Utilization Experience?
Authors: Akio Abe, Soichiro Motohashi, Yoshie Kadowaki
This poster presentation explores the potential of AI integration in the creation of Japanese proficiency tests. A hypothetical scenario was designed in which an unexpected situation required a Japanese language teacher to recreate an assessment test just one week before its administration. Two teachers participated: one with over ten years of Japanese language teaching experience but limited AI experience, and another with only six months of teaching experience but extensive experience in utilizing AI. Both were given a one-hour time limit to create a test based on the same instructional content.
The tests were then evaluated by another Japanese language teacher with extensive experience in both Japanese language education and AI usage, using a rubric-based assessment framework. The results showed that the test created by the teacher with significant AI experience, but limited teaching experience demonstrated greater content diversity and fewer minor errors in question formulation.
These findings suggest that while expertise in language education remains essential, AI utilization skills can also play a significant role in test creation. Moreover, they highlight the importance of further research on AI applications in teacher training and language education.
Weather and Climate in Modelling Energy Usage in Nevada
Authors: Kevin Bumgartner
There is a tight relationship between daily residential energy usage and the average daily temperature, with single-family homes in Las Vegas using up to roughly 4 times as much energy on very hot days as they use on more temperate days. For the past six years, the public utilities commission of Nevada has required energy utilities to incorporate local warming trends into the models used to compensate for temperature in long-term energy forecasting, and in models used for analysis of historical data used in rate-setting. Kevin Bumgartner's poster presents some representations of such models, along with some interesting plots of weather and usage data, the surprising results of distinct-month models of regional climate change in Reno and Las Vegas, and distinct-month models of the energy/temperature response. All modelling shown is done using linear regression.
Using R for Correspondence Analysis and Visualization of Multivariate Data: Relationships among Categorical Variables
Authors: Alexandria B. Stanton, Colleen I. Murray
Correspondence analysis (CAs) was conducted using R to examine associations between categorical variables without forcing data into a linear relationship. CA analysis tested associations among demographic variables and factors that may be associated with positive perceptions, such as hope.
Leveraging Machine Learning and NASA Satellite Data to Improve Wildfire Prediction and Environmental Insights
Authors: Michael Martin, Lei Yang Farnaz Hosseinpour
The increasing frequency of wildfires in California highlights the need for improved prediction methods to support environmental management. This project aims to improve wildfire prediction using NASA satellite remotely sensed big data. We analyzed 25 years of NASA high-resolution environmental variables over the Western U.S. to develop Machine Learning (ML) models for enhancing accuracy of fire-induced air pollutants. We developed deep learning architectures to enhance the precision of wildfire-related emission predictions. Feature importance analysis using SHAP improves model interpretability, ensuring transparency in AI-driven forecasts. This approach is crucial for developing efficient, explainable data-driven models to support informed decision-making in environmental scenarios.
Big Data in Environmental Science: Insights into Wildfire Smoke Emission
Authors: Rahele B. Vaezi, Farnaz Hosseinpour
Over the past few decades, there has been a significant rise in extreme fires, which have an adverse effect on air quality and the environment. Wildfire smoke can be transported over large distances, acting as air pollutants that affect adjacent and distant downwind communities. This study explored the possible relationships between wildfire smoke emissions and climate indices in the western U.S. Climatology helps identify long-term trends that rely on large-scale datasets. We applied an ensemble of remotely sensed observations and the historical Modern-Era Retrospective Analysis for Research and Applications, the second version (MERRA-2) reanalysis, and the Gridded Surface Meteorological (gridMET) data. These datasets were used due to their high spatial and temporal resolution, which provided insight into wildfire smoke transport and its effects on a regional scale. Given the global hourly MERRA-2 data since 1980, even a single variable can yield several terabytes of data over prolonged temporal scales, necessitating advanced computational methods, requiring specialized expertise for efficient storage, processing, and analysis.
Chess's Impact on Memory - A Study on the Predictive Power of Chess in Memory
Authors: Alec Brooks
This study explores the relationship between playing chess and memory improvement using data from a simulated population on The Islands website platform. Over a simulated fall semester, students were divided into control and participant groups, with the participants playing daily chess games while both groups completed daily memory tests. Although the participant group showed greater improvement in memory scores compared to the control group, the difference was low statistical significance, likely due to the small sample size. However, linear regression revealed that study duration and chess win/loss records were meaningful predictors of memory scores, accounting for a substantial portion of the observed variation. These findings suggest a potential link between playing chess and memory enhancement, but further research with larger, real-world populations is necessary to confirm these results.
Phylogenomic analysis sheds light on the origin of Burkholderia gladioli in Bangladesh
Authors: Ismam Ahmed Protic, Md. Nasir Uddin, Andrew Gorzalski, Md. Rashidul Islam, David Alvarez-Ponce
Bacterial Panicle Blight (BPB), caused by Burkholderia gladioli, has recently been found in multiple rice-growing regions in Bangladesh, raising questions about how the pathogen was introduced in the country. To investigate the introduction and origin of B. gladioli, we sequenced the complete genomes of 19 strains isolated from symptomatic rice panicles from four major rice-growing areas in Bangladesh. Construction of a phylogenetic tree using all 320 publicly available genomes along with the 19 newly sequenced genomes revealed 5 phylogenomically distinct clans. Our results suggest that B. gladioli was introduced in Bangladesh at least 5 times. The tree topology suggests that four of the introductions may have originated from the United States of America (USA), and these four clans are sister groups with several clinical strains of B. gladioli from the USA.
Wildlife and Weather Patterns: A Data-Driven Exploration
Authors: Seth Miller, Alec Brooks, Minsung Jung, Megan Lahti, Eriko Sakamura, Cecilia Vigil
From agriculture and infrastructure to energy and water supplies, climate plays a significant role in societal functionality and health. As weather and climate change grow increasingly volatile, it's essential to examine their impact on biodiversity. This project utilizes data gathered locally by faculty and students from the Biology Department at Truckee Meadows Community College (TMCC), analyzed in conjunction with weather, temperature, and snow accumulation data collected by the California Data Exchange Center (CDEC). This study aims to review the correlation between flowering data and seasonal weather conditions in the Great Basic region. This is a part of TMCC Undergraduate research project mentored by four TMCC faculties and supported by INBRE and NASA grants.
Gym Member Activity
Authors: Ty Martin
Many gym-goers either lack structured workout knowledge or are experienced individuals seeking optimization. This study analyzes the relationship between workout types, water intake, and exercise duration using the Gym Members Exercise Dataset from Kaggle. Key research questions focus on whether workout type affects BMI, if water intake influences session duration, and whether workout frequency correlates with body weight. Data was categorized by workout type, frequency, session duration, and gender. Findings indicate no significant difference in workout preference between genders. Additionally, increased water consumption led to shorter workout durations, while frequent exercise correlated with lower or recommended body weight. However, workout type did not significantly impact BMI. These insights contribute to a better understanding of exercise habits and their effects on fitness outcomes.
The Bridge to AI and Neurocomputing Core of the UNR Center for Integrative Neuroscience
Authors: Michael Rudd, Alireza Tavakkoli, Eelke Folmer, Michael A. Webster
The new Bridge to AI and Neurocomputing (BAIN) Core of the UNR Center for Integrative Neuroscience provides expert consulting in the areas of machine learning and AI, advanced statistical methods and computational modeling, and AR/VR and visualization to help investigators apply the latest computational modeling methods in their research. Our machine learning experts can help you pose your research hypotheses in the form of testable algorithmic models and run these models on our high-speed online pipeline. Our statistics consultant can help you prepare methods sections for grant submissions or publications, revise manuscripts, and advise you on experimental design, experimental design, and power analysis. We also offer expertise and equipment for applying virtual and augmented reality in your research. Come to this poster to find out how these methods can be used to enhance your research, and join the computational modeling and AI revolution today!