Yiwen Liu

Interests

Research

Big data analytics, statistical learning for high-dimensional data, multiple sources data fusion, and bioinformatics.

Courses

Scholarly Contributions

Chapters

• Liu, Y., & Xie, R. (2025). Common issues in analysis. In Translational Gastroenterology: Handbook for Designing and Conducting Clinical and Translational Research. Elsevier. doi:10.1016/b978-0-12-821426-8.00079-0
  More info

  The use of statistical methods is a key component in biomedical research. It allows researchers to draw reasonable conclusions and make valid inferences in the general population. Statistics plays an important role not only in the design stage of a clinical study but also in the data collection, data analysis, and reporting stages. A sound understanding of statistical concepts and reasonings is of primary importance. In this chapter, we discussed several common issues regarding analyzing medical data as well as corresponding strategies in dealing with these issues.

Journals/Publications

• Liu, T., Hollister, J., Kern, K. J., Valenti, M., Beitel, S. C., Gulotta, J. J., Jahnke, S. A., Buren, H., Haseney, J., O'Neill, B., St Clair, C., Liu, Y., von Hippel, F., Mullins, C. E., Walker, D. I., Goodrich, J. M., Burgess, J. L., & Furlong, M. A. (2026). Differential metabolic profiles by training fire exposure in female firefighters. International journal of hygiene and environmental health, 273, 114746.
  More info

  Female firefighters face elevated risks for cancer and reproductive disorders, but the underlying metabolic mechanisms remain unclear.
• Liu, T., Furlong, M. A., Snider, J. M., Beitel, S., Mullins, C. E., Walker, D. I., Goodrich, J. M., Urwin, D. J., Gabriel, J., Hughes, J., Gulotta, J. J., Calkins, M. M., Liu, Y., von Hippel, F. A., Beamer, P., & Burgess, J. L. (2025). Evaluating urinary metabolic profiles with wildland-urban-interface (wui) fire exposure among male firefighters: a comparison with municipal structure fires (msf). Environmental health : a global access science source, 24(1), 88.
  More info

  Firefighters have frequent exposure to carcinogens and an increased risk of cancer. Wildland-urban interface (WUI) fires, which involve both structures and undeveloped wildland fuels, pose unique challenges to the health of firefighters. However, the extent of health risks associated with these fires remains underexplored.
• Valenti, M. A., Farland, L. V., Huang, K., Liu, Y., Beitel, S. C., Jahnke, S. A., Hollerbach, B., St Clair, C. C., Gulotta, J. J., Kolar, J. J., Urwin, D. J., Louzado-Feliciano, P., Baker, J. B., Jack, K. L., Caban-Martinez, A. J., Goodrich, J. M., & Burgess, J. L. (2025). Evaluating the Effect of Depression, Anxiety, and Post-Traumatic Stress Disorder on Anti-Müllerian Hormone Levels Among Women Firefighters. Journal of women's health (2002), 34(3), 354-361.
  More info

  To assess whether depression, anxiety, and post-traumatic stress disorder (PTSD) are associated with serum anti-Müllerian hormone (AMH) levels. We used data from a sample of women firefighters from the Fire Fighter Cancer Cohort Study. Participant demographics, reproductive history, and self-reported clinical diagnosis of anxiety, depression, and PTSD were collected with serum for AMH analysis at enrollment. Linear regression models were used to estimate the association between anxiety, depression, and PTSD and log transformed AMH levels adjusted for age years (continuous and squared) and body mass index. Percent difference in AMH was calculated by [exp(β) - 1] × 100. Among 372 participants, with mean ± standard deviation age 32.54 ± 6.32, clinical diagnoses were reported as follows: depression (15%), anxiety (18.2%), or PTSD (8.7%). No statistically significant association was observed between depression and AMH levels (-22%Δ, 95% confidence interval [CI]: -47.3, 14.5). Women firefighters with a history of anxiety (-33%Δ, 95% CI: -53.5, -4.2) and PTSD (-66%Δ, 95% CI: -79.1, -44.6) had lower serum AMH compared with participants without those conditions. When individuals with concurrent PTSD were excluded, the association between anxiety ceased to be statistically significant (26.7%Δ, 95% CI: -17.9, 92.6). A history of clinically diagnosed anxiety or PTSD was associated with statistically significantly lower AMH levels. This association offers insight into the potential biological mechanisms through which mental health conditions may influence reproductive health.
• Valenti, M. A., Farland, L. V., Huang, K., Liu, Y., Beitel, S. C., Jahnke, S. A., Hollerbach, B., St. Clair, C. C., Gulotta, J. J., Kolar, J. J., Urwin, D. J., Louzado-Feliciano, P., Baker, J. B., Jack, K. L., Caban-Martinez, A. J., Goodrich, J. M., & Burgess, J. L. (2025). Evaluating the Effect of Depression, Anxiety, and Post-Traumatic Stress Disorder on Anti-Müllerian Hormone Levels Among Women Firefighters. Journal of Women's Health, 34(Issue 3). doi:10.1089/jwh.2024.0534
  More info

  Objective: To assess whether depression, anxiety, and post-traumatic stress disorder (PTSD) are associated with serum anti-Müllerian hormone (AMH) levels. Study Design: We used data from a sample of women firefighters from the Fire Fighter Cancer Cohort Study. Participant demographics, reproductive history, and self-reported clinical diagnosis of anxiety, depression, and PTSD were collected with serum for AMH analysis at enrollment. Main Outcome Measure: Linear regression models were used to estimate the association between anxiety, depression, and PTSD and log transformed AMH levels adjusted for age years (continuous and squared) and body mass index. Percent difference in AMH was calculated by [exp(β) - 1] × 100. Results: Among 372 participants, with mean ± standard deviation age 32.54 ± 6.32, clinical diagnoses were reported as follows: depression (15%), anxiety (18.2%), or PTSD (8.7%). No statistically significant association was observed between depression and AMH levels (−22%Δ, 95% confidence interval [CI]: −47.3, 14.5). Women firefighters with a history of anxiety (−33%Δ, 95% CI: −53.5, −4.2) and PTSD (−66%Δ, 95% CI: −79.1, −44.6) had lower serum AMH compared with participants without those conditions. When individuals with concurrent PTSD were excluded, the association between anxiety ceased to be statistically significant (26.7%Δ, 95% CI: −17.9, 92.6). Conclusion: A history of clinically diagnosed anxiety or PTSD was associated with statistically significantly lower AMH levels. This association offers insight into the potential biological mechanisms through which mental health conditions may influence reproductive health.
• Zhang, M., Parker, J., An, L., Liu, Y., & Sun, X. (2025). Flexible analysis of spatial transcriptomics data (FAST): a deconvolution approach. BMC Bioinformatics, 26(1), 35.
• Sun, X., Liu, Y., Zhong, W., & Li, B. (2022). B-scaling: A novel nonparametric data fusion method. The Annals of Applied Statistics, 16(3). doi:10.1214/21-aoas1537
• Liu, Y., Zhong, W., & Zeng, P. (2021). A Model-free Variable Screening Method Based on Leverage Score. Journal of the American Statistical Association, 1-12. doi:10.1080/01621459.2021.1918554
• Zhang, M., Liu, Y., Zhou, H., Watkins, J., & Zhou, J. (2021). A novel nonlinear dimension reduction approach to infer population structure for low-coverage sequencing data. BMC bioinformatics, 22(1). doi:10.1186/s12859-021-04265-7
  More info

  BACKGROUND: Low-depth sequencing allows researchers to increase sample size at the expense of lower accuracy. To incorporate uncertainties while maintaining statistical power, we introduce MCPCA_PopGen to analyze population structure of low-depth sequencing data. RESULTS: The method optimizes the choice of nonlinear transformations of dosages to maximize the Ky Fan norm of the covariance matrix. The transformation incorporates the uncertainty in calling between heterozygotes and the common homozygotes for loci having a rare allele and is more linear when both variants are common. CONCLUSIONS: We apply MCPCA_PopGen to samples from two indigenous Siberian populations and reveal hidden population structure accurately using only a single chromosome. The MCPCA_PopGen package is available on https://github.com/yiwenstat/MCPCA_PopGen .
• Sun, X., Liu, Y., & An, L. (2020). EDGE: Ensemble Dimensionality Reduction and Feature Gene Extraction for Single-cell RNA-seq Data. Nature Communications.

Presentations

• Liu, Y. (2019, May). B-scaling: a novel nonparametric data fusion method. New England Statistics Symposium.
• Liu, Y. (2019, October). Trajectory inference using single-cell transcriptomics data. TRIPODS RWG 6.
• Liu, Y. (2018, Dec). B-scaling: a novel nonparametric data fusion method. International Conference on Big Data and Information Analytics. Houston, TX.
• Liu, Y. (2018, Sep). Statistical leverage and its usage in variable screening. Department seminar, Department of Epidemiology and Biostatistics. Tucson, AZ.
• Liu, Y., Watkins, J., & Encinas, A. (2018, Sep). Sodium Channels Pathologies and Statistical Issues in Pathogenicity Prediction. Quantitative Biology Colloquium. Tucson, AZ: Department of Mathematics.

Others

• Liu, Y. (2018, Jun). Statistical learning for high-dimensional and complex data session chair. ICSA Applied Statistics Symposium.