Nazia Riasat
Tagline:Ph.D. Candidate in Statistics, North Dakota State University
Fargo, ND, United States
About & Research Focus
I am a Ph.D. candidate in Statistics at North Dakota State University, studying how mathematical models can capture the complexity of biological systems. My research combines statistical rigor with biological insight to build interpretable and reproducible frameworks for genomic data analysis. I am particularly interested in developing tools that promote transparency, reliability, and collaboration across data science and biomedical research.
My current work focuses on benchmarking simulation methods for bulk RNA-seq data across diverse biological settings. Using publicly available transcriptomic datasets, I generate synthetic data calibrated to real experimental profiles and systematically evaluate widely used simulators based on their ability to reproduce key statistical properties and biological structure. This evaluation leverages a comprehensive set of distributional, correlation-based, and multivariate metrics to assess realism, identify strengths and limitations of existing tools, and highlight gaps in modeling biological variability.
Building on these insights, my dissertation advances a new RNA-seq simulation framework designed to better capture complex expression dependence, heterogeneity, and multivariate structure observed in real data. The method integrates flexible statistical modeling with correlation-preserving mechanisms to improve biological fidelity while maintaining computational efficiency.
Work Experiences
Publications
When Stability Fails: Hidden Failure Modes of LLMS in Data-Constrained Scientific Decision-Making
Conference PaperDate:2026Authors:Description:ICLR 2026 Workshop: I Can’t Believe It’s Not Better (ICBINB)
Proceedings of Machine Learning Research (PMLR), 2026Incorporating Tobacco/Nicotine Dependence Treatment Education Into a Nurse Practitioner Program
Journal ArticlePublisher:The Journal for Nurse PractitionersDate:2025Authors:Allison PeltierMykell BarnacleNazia RiasatMegan OrrKanchan BhattaraiJillian DoanKelly Buettner-SchmidtAn exploration of graph distances, graph curvature, and applications to network analysis
Book ChapterPublisher:SpringerNatureDate:2025Authors:Kasia Jankiewicz1Manasa Kesapragada1Anna Konstorum2Kathryn Leonard3Nazia Riasat4and Michelle Snider2
Projects
Public Health and Biostatistics
date: 2024Organization:North Dakota State University
Description:I conducted statistical analyses integrating evidence-based tobacco and nicotine dependence treatment into advanced nursing curricula. This project validated significant improvements in nurse practitioner students’ knowledge, confidence, and counseling ability.
Vaccine Hesitancy Interventions
date: 2022Organization:North Dakota State University
Description:During the COVID-19 pandemic, I led statistical analyses of the survey of healthcare students in addressing vaccine hesitancy. Results showed statistically significant gains in knowledge, confidence, and a measurable reduction in vaccine-hesitant attitudes. Importantly, 92% of participants reported intent to apply these strategies with patients.
Pediatric Obesity Outcomes
date: 2022Organization:North Dakota State University
Description:I analyzed over 230,000 pediatric health visits in an NIH-funded study of obesity prevalence before, during, and after the pandemic. Results revealed significant, persistent increases in obesity, particularly among adolescents, rural populations, and American Indian/Alaska Native children even after return to pre-pandemic activities. My analyses and visualizations highlighted the role of social determinants of health in exacerbating disparities, providing critical evidence for designing targeted federal and state interventions.
Transportation Safety for Teen Drivers
date: 2021Organization:North Dakota State University
Description:I led the statistical analysis and writing on the safety outcomes of 15-year-old novice drivers under Graduated Driver Licensing policy. Using logistic regression on over 15,000 records, I demonstrated that public school driver education programs reduced crash risk by 32%. This evidence supports policy reforms aligned with federal “Vision Zero” goals to eliminate roadway fatalities.