A non-negative spike-and-slab lasso generalized linear stacking prediction modeling method for high-dimensional omics data
Abstract Background High-dimensional omics data are increasingly utilized in clinical and public health research for disease risk prediction. Many previous sparse methods have been proposed that using prior knowledge, e.g., biological group structure information, to guide the model-building process....
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2024-03-01
|
| Series: | BMC Bioinformatics |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12859-024-05741-6 |
| _version_ | 1797247009553383424 |
|---|---|
| author | Junjie Shen Shuo Wang Yongfei Dong Hao Sun Xichao Wang Zaixiang Tang |
| author_facet | Junjie Shen Shuo Wang Yongfei Dong Hao Sun Xichao Wang Zaixiang Tang |
| author_sort | Junjie Shen |
| collection | DOAJ |
| description | Abstract Background High-dimensional omics data are increasingly utilized in clinical and public health research for disease risk prediction. Many previous sparse methods have been proposed that using prior knowledge, e.g., biological group structure information, to guide the model-building process. However, these methods are still based on a single model, offen leading to overconfident inferences and inferior generalization. Results We proposed a novel stacking strategy based on a non-negative spike-and-slab Lasso (nsslasso) generalized linear model (GLM) for disease risk prediction in the context of high-dimensional omics data. Briefly, we used prior biological knowledge to segment omics data into a set of sub-data. Each sub-model was trained separately using the features from the group via a proper base learner. Then, the predictions of sub-models were ensembled by a super learner using nsslasso GLM. The proposed method was compared to several competitors, such as the Lasso, grlasso, and gsslasso, using simulated data and two open-access breast cancer data. As a result, the proposed method showed robustly superior prediction performance to the optimal single-model method in high-noise simulated data and real-world data. Furthermore, compared to the traditional stacking method, the proposed nsslasso stacking method can efficiently handle redundant sub-models and identify important sub-models. Conclusions The proposed nsslasso method demonstrated favorable predictive accuracy, stability, and biological interpretability. Additionally, the proposed method can also be used to detect new biomarkers and key group structures. |
| first_indexed | 2024-04-24T19:51:53Z |
| format | Article |
| id | doaj.art-935b7147842b49e2aa08ed2f83118512 |
| institution | Directory Open Access Journal |
| issn | 1471-2105 |
| language | English |
| last_indexed | 2024-04-24T19:51:53Z |
| publishDate | 2024-03-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Bioinformatics |
| spelling | doaj.art-935b7147842b49e2aa08ed2f831185122024-03-24T12:35:30ZengBMCBMC Bioinformatics1471-21052024-03-0125112010.1186/s12859-024-05741-6A non-negative spike-and-slab lasso generalized linear stacking prediction modeling method for high-dimensional omics dataJunjie Shen0Shuo Wang1Yongfei Dong2Hao Sun3Xichao Wang4Zaixiang Tang5Department of Biostatistics, School of Public Health, Jiangsu Key Laboratory of Preventive and Translational Medicine for Geriatric Diseases, MOE Key Laboratory of Geriatric Diseases and Immunology, Suzhou Medical College of Soochow UniversityInstitute of Medical Biometry and Statistics, Faculty of Medicine and Medical Center, University of FreiburgDepartment of Biostatistics, School of Public Health, Jiangsu Key Laboratory of Preventive and Translational Medicine for Geriatric Diseases, MOE Key Laboratory of Geriatric Diseases and Immunology, Suzhou Medical College of Soochow UniversityDepartment of Biostatistics, School of Public Health, Jiangsu Key Laboratory of Preventive and Translational Medicine for Geriatric Diseases, MOE Key Laboratory of Geriatric Diseases and Immunology, Suzhou Medical College of Soochow UniversityDepartment of Biostatistics, School of Public Health, Jiangsu Key Laboratory of Preventive and Translational Medicine for Geriatric Diseases, MOE Key Laboratory of Geriatric Diseases and Immunology, Suzhou Medical College of Soochow UniversityDepartment of Biostatistics, School of Public Health, Jiangsu Key Laboratory of Preventive and Translational Medicine for Geriatric Diseases, MOE Key Laboratory of Geriatric Diseases and Immunology, Suzhou Medical College of Soochow UniversityAbstract Background High-dimensional omics data are increasingly utilized in clinical and public health research for disease risk prediction. Many previous sparse methods have been proposed that using prior knowledge, e.g., biological group structure information, to guide the model-building process. However, these methods are still based on a single model, offen leading to overconfident inferences and inferior generalization. Results We proposed a novel stacking strategy based on a non-negative spike-and-slab Lasso (nsslasso) generalized linear model (GLM) for disease risk prediction in the context of high-dimensional omics data. Briefly, we used prior biological knowledge to segment omics data into a set of sub-data. Each sub-model was trained separately using the features from the group via a proper base learner. Then, the predictions of sub-models were ensembled by a super learner using nsslasso GLM. The proposed method was compared to several competitors, such as the Lasso, grlasso, and gsslasso, using simulated data and two open-access breast cancer data. As a result, the proposed method showed robustly superior prediction performance to the optimal single-model method in high-noise simulated data and real-world data. Furthermore, compared to the traditional stacking method, the proposed nsslasso stacking method can efficiently handle redundant sub-models and identify important sub-models. Conclusions The proposed nsslasso method demonstrated favorable predictive accuracy, stability, and biological interpretability. Additionally, the proposed method can also be used to detect new biomarkers and key group structures.https://doi.org/10.1186/s12859-024-05741-6Stacking Bayesian methodNon-negative spike-and-slab priorOmics segmentation |
| spellingShingle | Junjie Shen Shuo Wang Yongfei Dong Hao Sun Xichao Wang Zaixiang Tang A non-negative spike-and-slab lasso generalized linear stacking prediction modeling method for high-dimensional omics data BMC Bioinformatics Stacking Bayesian method Non-negative spike-and-slab prior Omics segmentation |
| title | A non-negative spike-and-slab lasso generalized linear stacking prediction modeling method for high-dimensional omics data |
| title_full | A non-negative spike-and-slab lasso generalized linear stacking prediction modeling method for high-dimensional omics data |
| title_fullStr | A non-negative spike-and-slab lasso generalized linear stacking prediction modeling method for high-dimensional omics data |
| title_full_unstemmed | A non-negative spike-and-slab lasso generalized linear stacking prediction modeling method for high-dimensional omics data |
| title_short | A non-negative spike-and-slab lasso generalized linear stacking prediction modeling method for high-dimensional omics data |
| title_sort | non negative spike and slab lasso generalized linear stacking prediction modeling method for high dimensional omics data |
| topic | Stacking Bayesian method Non-negative spike-and-slab prior Omics segmentation |
| url | https://doi.org/10.1186/s12859-024-05741-6 |
| work_keys_str_mv | AT junjieshen anonnegativespikeandslablassogeneralizedlinearstackingpredictionmodelingmethodforhighdimensionalomicsdata AT shuowang anonnegativespikeandslablassogeneralizedlinearstackingpredictionmodelingmethodforhighdimensionalomicsdata AT yongfeidong anonnegativespikeandslablassogeneralizedlinearstackingpredictionmodelingmethodforhighdimensionalomicsdata AT haosun anonnegativespikeandslablassogeneralizedlinearstackingpredictionmodelingmethodforhighdimensionalomicsdata AT xichaowang anonnegativespikeandslablassogeneralizedlinearstackingpredictionmodelingmethodforhighdimensionalomicsdata AT zaixiangtang anonnegativespikeandslablassogeneralizedlinearstackingpredictionmodelingmethodforhighdimensionalomicsdata AT junjieshen nonnegativespikeandslablassogeneralizedlinearstackingpredictionmodelingmethodforhighdimensionalomicsdata AT shuowang nonnegativespikeandslablassogeneralizedlinearstackingpredictionmodelingmethodforhighdimensionalomicsdata AT yongfeidong nonnegativespikeandslablassogeneralizedlinearstackingpredictionmodelingmethodforhighdimensionalomicsdata AT haosun nonnegativespikeandslablassogeneralizedlinearstackingpredictionmodelingmethodforhighdimensionalomicsdata AT xichaowang nonnegativespikeandslablassogeneralizedlinearstackingpredictionmodelingmethodforhighdimensionalomicsdata AT zaixiangtang nonnegativespikeandslablassogeneralizedlinearstackingpredictionmodelingmethodforhighdimensionalomicsdata |