Causality-driven candidate identification for reliable DNA methylation biomarker discovery
Abstract Despite vast data support in DNA methylation (DNAm) biomarker discovery to facilitate health-care research, this field faces huge resource barriers due to preliminary unreliable candidates and the consequent compensations using expensive experiments. The underlying challenges lie in the con...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-025-56054-y |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832594567714570240 |
|---|---|
| author | Xinlu Tang Rui Guo Zhanfeng Mo Wenli Fu Xiaohua Qian |
| author_facet | Xinlu Tang Rui Guo Zhanfeng Mo Wenli Fu Xiaohua Qian |
| author_sort | Xinlu Tang |
| collection | DOAJ |
| description | Abstract Despite vast data support in DNA methylation (DNAm) biomarker discovery to facilitate health-care research, this field faces huge resource barriers due to preliminary unreliable candidates and the consequent compensations using expensive experiments. The underlying challenges lie in the confounding factors, especially measurement noise and individual characteristics. To achieve reliable identification of a candidate pool for DNAm biomarker discovery, we propose a Causality-driven Deep Regularization framework to reinforce correlations that are suggestive of causality with disease. It integrates causal thinking, deep learning, and biological priors to handle non-causal confounding factors, through a contrastive scheme and a spatial-relation regularization that reduces interferences from individual characteristics and noises, respectively. The comprehensive reliability of the proposed method was verified by simulations and applications involving various human diseases, sample origins, and sequencing technologies, highlighting its universal biomedical significance. Overall, this study offers a causal-deep-learning-based perspective with a compatible tool to identify reliable DNAm biomarker candidates, promoting resource-efficient biomarker discovery. |
| format | Article |
| id | doaj-art-b4764d2ce2e34f19a63ed6c2e777fa4c |
| institution | Kabale University |
| issn | 2041-1723 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj-art-b4764d2ce2e34f19a63ed6c2e777fa4c2025-01-19T12:31:28ZengNature PortfolioNature Communications2041-17232025-01-0116111510.1038/s41467-025-56054-yCausality-driven candidate identification for reliable DNA methylation biomarker discoveryXinlu Tang0Rui Guo1Zhanfeng Mo2Wenli Fu3Xiaohua Qian4The Medical Image and Health Informatics Lab, the School of Biomedical Engineering, Shanghai Jiao Tong UniversityThe Medical Image and Health Informatics Lab, the School of Biomedical Engineering, Shanghai Jiao Tong UniversityCollege of Computing and Data Science, Nanyang Technological UniversityThe Medical Image and Health Informatics Lab, the School of Biomedical Engineering, Shanghai Jiao Tong UniversityThe Medical Image and Health Informatics Lab, the School of Biomedical Engineering, Shanghai Jiao Tong UniversityAbstract Despite vast data support in DNA methylation (DNAm) biomarker discovery to facilitate health-care research, this field faces huge resource barriers due to preliminary unreliable candidates and the consequent compensations using expensive experiments. The underlying challenges lie in the confounding factors, especially measurement noise and individual characteristics. To achieve reliable identification of a candidate pool for DNAm biomarker discovery, we propose a Causality-driven Deep Regularization framework to reinforce correlations that are suggestive of causality with disease. It integrates causal thinking, deep learning, and biological priors to handle non-causal confounding factors, through a contrastive scheme and a spatial-relation regularization that reduces interferences from individual characteristics and noises, respectively. The comprehensive reliability of the proposed method was verified by simulations and applications involving various human diseases, sample origins, and sequencing technologies, highlighting its universal biomedical significance. Overall, this study offers a causal-deep-learning-based perspective with a compatible tool to identify reliable DNAm biomarker candidates, promoting resource-efficient biomarker discovery.https://doi.org/10.1038/s41467-025-56054-y |
| spellingShingle | Xinlu Tang Rui Guo Zhanfeng Mo Wenli Fu Xiaohua Qian Causality-driven candidate identification for reliable DNA methylation biomarker discovery Nature Communications |
| title | Causality-driven candidate identification for reliable DNA methylation biomarker discovery |
| title_full | Causality-driven candidate identification for reliable DNA methylation biomarker discovery |
| title_fullStr | Causality-driven candidate identification for reliable DNA methylation biomarker discovery |
| title_full_unstemmed | Causality-driven candidate identification for reliable DNA methylation biomarker discovery |
| title_short | Causality-driven candidate identification for reliable DNA methylation biomarker discovery |
| title_sort | causality driven candidate identification for reliable dna methylation biomarker discovery |
| url | https://doi.org/10.1038/s41467-025-56054-y |
| work_keys_str_mv | AT xinlutang causalitydrivencandidateidentificationforreliablednamethylationbiomarkerdiscovery AT ruiguo causalitydrivencandidateidentificationforreliablednamethylationbiomarkerdiscovery AT zhanfengmo causalitydrivencandidateidentificationforreliablednamethylationbiomarkerdiscovery AT wenlifu causalitydrivencandidateidentificationforreliablednamethylationbiomarkerdiscovery AT xiaohuaqian causalitydrivencandidateidentificationforreliablednamethylationbiomarkerdiscovery |