Bootstrap-enhanced regularization addressing multicollinearity and skewness in high-dimensional immunophenotyping data.
8 affiliations
- 1Department of Public Health Sciences, College of Dentistry, Texas A&M University, Dallas, TX, 75246, USA.
- 2Department of Biomedical Sciences, College of Dentistry, Texas A&M University, Dallas, TX, 75246, USA.
- 3Department of Biostatistics, College of Public Health and Health Professions, University of Florida, Gainesville, FL, 32610, USA.
- 4Diabetes Institute, University of Florida, Gainesville, FL, 32610, USA.
- 5Department of Biostatistics, University of Texas MD Anderson Cancer Center, Houston, TX, 77030, USA.
- 6Department of Pathology, Immunology, and Laboratory Medicine, College of Medicine, University of Florida, Gainesville, FL, 32610, USA.
- 7Department of Biostatistics, College of Public Health and Health Professions, University of Florida, Gainesville, FL, 32610, USA. [email protected].
- 8Diabetes Institute, University of Florida, Gainesville, FL, 32610, USA. [email protected].
Abstract
Background
Accurate identification and estimation of variables associated with outcomes or disease states are critical for advancing diagnosis, prognosis, and precision medicine in biomedical research. Regularized regression techniques, such as lasso, are widely employed to enhance interpretability by reducing model complexity and identifying significant variables. However, these methods face two major challenges: (1) the exclusion of important variables due to high correlation with included predictors, and (2) the presence of skewness in human biomedical datasets, which violates key statistical assumptions. Current approaches that fail to address these issues simultaneously may lead to biased interpretations and unreliable coefficient estimates. To overcome these limitations, we propose an enhanced two-step approach, the Bootstrap-Enhanced Regularization Method (BERM).Results
BERM outperformed existing regularization methods in variable selection, achieving the highest overall balanced accuracy while maintaining competitive coefficient estimation performance across a range of simulated sparsity, noise, and dimensionality scenarios. We further demonstrated the effectiveness of BERM by applying it to a human immunophenotyping dataset to identify important immune parameters in the autoimmune disease, type 1 diabetes.Conclusion
BERM is a robust approach for variable selection and coefficient estimation in complex biomedical datasets. Its consistent performance across a wide range of data conditions supports more reliable identification of important variables. An open-source implementation of BERM is available as an R package on GitHub ( https://github.com/xiaorudong/berm ).
Related
- Strict non-tightness of the level-3 quantum bootstrap for a natural three-dimensional Hamiltonian, at an excited level
- Adaptable High-Dimensional Change Point Detection via Ridge Regularization
- High-Dimensional Regularization of the Spatial Sign Covariance Matrix for Robust Shape Estimation
- Functional Dimensional Regularization
- Multicollinearity-agnostic feature screening for non-Euclidean responses: a factor adjusted approach
BibTeX
@article{dong2026bootstrapenhanced,
title = {{Bootstrap-enhanced regularization addressing multicollinearity and skewness in high-dimensional immunophenotyping data.}},
author = {Xiaoru Dong and Apoorva Goyal and Muxuan Liang and Maigan A Brusko and Todd M Brusko and Rhonda Bacher},
journal = {BMC bioinformatics},
year = {2026},
month = aug,
doi = {10.1186/s12859-026-06586-x},
url = {https://europepmc.org/article/MED/42972258},
note = {CC BY},
}
RIS
TY - JOUR
TI - Bootstrap-enhanced regularization addressing multicollinearity and skewness in high-dimensional immunophenotyping data.
AU - Xiaoru Dong
AU - Apoorva Goyal
AU - Muxuan Liang
AU - Maigan A Brusko
AU - Todd M Brusko
AU - Rhonda Bacher
PY - 2026
DA - 2026/08/06
JO - BMC bioinformatics
DO - 10.1186/s12859-026-06586-x
UR - https://europepmc.org/article/MED/42972258
AB - Background Accurate identification and estimation of variables associated with outcomes or disease states are critical for advancing diagnosis, prognosis, and precision medicine in biomedical research. Regularized regression techniques, such as lasso, are widely employed to enhance interpretability by reducing model complexity and identifying significant variables. However, these methods face two major challenges: (1) the exclusion of important variables due to high correlation with included predictors, and (2) the presence of skewness in human biomedical datasets, which violates key statistical assumptions. Current approaches that fail to address these issues simultaneously may lead to biased interpretations and unreliable coefficient estimates. To overcome these limitations, we propose an enhanced two-step approach, the Bootstrap-Enhanced Regularization Method (BERM).Results BERM outperformed existing regularization methods in variable selection, achieving the highest overall balanced accuracy while maintaining competitive coefficient estimation performance across a range of simulated sparsity, noise, and dimensionality scenarios. We further demonstrated the effectiveness of BERM by applying it to a human immunophenotyping dataset to identify important immune parameters in the autoimmune disease, type 1 diabetes.Conclusion BERM is a robust approach for variable selection and coefficient estimation in complex biomedical datasets. Its consistent performance across a wide range of data conditions supports more reliable identification of important variables. An open-source implementation of BERM is available as an R package on GitHub ( https://github.com/xiaorudong/berm ).
SN - 1471-2105
ER -