Deep fusion of incomplete multi-omic data for molecular mechanism of Alzheimer’s disease
Abstract Multi-omics data provides a comprehensive view of biological systems and enables researchers to uncover intricate molecular mechanisms underlying complex diseases. However, multi-omic data is often incomplete and joint modeling of multi-omics data will lead to exclusion of a large portion o...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-08-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-14636-2 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849234286223294464 |
|---|---|
| author | Linhui Xie Yash Raj Mingzhao Tong Kwangsik Nho Paul Salama Andrew J. Saykin Shiaofen Fang Jingwen Yan |
| author_facet | Linhui Xie Yash Raj Mingzhao Tong Kwangsik Nho Paul Salama Andrew J. Saykin Shiaofen Fang Jingwen Yan |
| author_sort | Linhui Xie |
| collection | DOAJ |
| description | Abstract Multi-omics data provides a comprehensive view of biological systems and enables researchers to uncover intricate molecular mechanisms underlying complex diseases. However, multi-omic data is often incomplete and joint modeling of multi-omics data will lead to exclusion of a large portion of subjects. Furthermore, most current multi-omics studies pinpoint individual -omics markers, which may not interact, posing challenges for interpretation. In this study, we developed an interpretable deep trans-omic fusion neural network, TransFuse, to include incomplete -omic data for training of prediction models. When evaluated using the data from two Alzheimer’s disease cohorts, TransFuse generally showed superior or comparable performance over competing methods in a wide range of metrics like classification accuracy and F1 score. In addition, TransFuse yielded a subset of multi-omics features forming functional disease network modules, providing valuable insights into underlying molecular mechanism. In addition, almost all the genetic variants identified by TransFuse are expression quantitative trait locus (eQTLs) specific to frontal cortex tissue, from which the gene and protein expression data were collected. This highlights the great potential of TransFuse in capturing the tissue-specific information flow. Top pathways enriched include VEGF and EPH pathways, both influencing neural development and synaptic formation. |
| format | Article |
| id | doaj-art-0e13516ee17d4b2fa8b6c07e77a693d8 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-0e13516ee17d4b2fa8b6c07e77a693d82025-08-20T04:03:12ZengNature PortfolioScientific Reports2045-23222025-08-0115111110.1038/s41598-025-14636-2Deep fusion of incomplete multi-omic data for molecular mechanism of Alzheimer’s diseaseLinhui Xie0Yash Raj1Mingzhao Tong2Kwangsik Nho3Paul Salama4Andrew J. Saykin5Shiaofen Fang6Jingwen Yan7Purdue University, Indianapolis, Department of Electrical and Computer EngineeringIndiana University, Indianapolis, Department of Biomedical Engineering and InformaticsIndiana University, Indianapolis, Department of Computer ScienceIndiana University School of Medicine, Department of Radiology and Imaging SciencesPurdue University, Indianapolis, Department of Electrical and Computer EngineeringIndiana University School of Medicine, Department of Radiology and Imaging SciencesIndiana University, Indianapolis, Department of Computer ScienceIndiana University, Indianapolis, Department of Biomedical Engineering and InformaticsAbstract Multi-omics data provides a comprehensive view of biological systems and enables researchers to uncover intricate molecular mechanisms underlying complex diseases. However, multi-omic data is often incomplete and joint modeling of multi-omics data will lead to exclusion of a large portion of subjects. Furthermore, most current multi-omics studies pinpoint individual -omics markers, which may not interact, posing challenges for interpretation. In this study, we developed an interpretable deep trans-omic fusion neural network, TransFuse, to include incomplete -omic data for training of prediction models. When evaluated using the data from two Alzheimer’s disease cohorts, TransFuse generally showed superior or comparable performance over competing methods in a wide range of metrics like classification accuracy and F1 score. In addition, TransFuse yielded a subset of multi-omics features forming functional disease network modules, providing valuable insights into underlying molecular mechanism. In addition, almost all the genetic variants identified by TransFuse are expression quantitative trait locus (eQTLs) specific to frontal cortex tissue, from which the gene and protein expression data were collected. This highlights the great potential of TransFuse in capturing the tissue-specific information flow. Top pathways enriched include VEGF and EPH pathways, both influencing neural development and synaptic formation.https://doi.org/10.1038/s41598-025-14636-2Transfer learningMulti-omic networkAlzheimer’s DiseaseSystems biologyDeep learning |
| spellingShingle | Linhui Xie Yash Raj Mingzhao Tong Kwangsik Nho Paul Salama Andrew J. Saykin Shiaofen Fang Jingwen Yan Deep fusion of incomplete multi-omic data for molecular mechanism of Alzheimer’s disease Scientific Reports Transfer learning Multi-omic network Alzheimer’s Disease Systems biology Deep learning |
| title | Deep fusion of incomplete multi-omic data for molecular mechanism of Alzheimer’s disease |
| title_full | Deep fusion of incomplete multi-omic data for molecular mechanism of Alzheimer’s disease |
| title_fullStr | Deep fusion of incomplete multi-omic data for molecular mechanism of Alzheimer’s disease |
| title_full_unstemmed | Deep fusion of incomplete multi-omic data for molecular mechanism of Alzheimer’s disease |
| title_short | Deep fusion of incomplete multi-omic data for molecular mechanism of Alzheimer’s disease |
| title_sort | deep fusion of incomplete multi omic data for molecular mechanism of alzheimer s disease |
| topic | Transfer learning Multi-omic network Alzheimer’s Disease Systems biology Deep learning |
| url | https://doi.org/10.1038/s41598-025-14636-2 |
| work_keys_str_mv | AT linhuixie deepfusionofincompletemultiomicdataformolecularmechanismofalzheimersdisease AT yashraj deepfusionofincompletemultiomicdataformolecularmechanismofalzheimersdisease AT mingzhaotong deepfusionofincompletemultiomicdataformolecularmechanismofalzheimersdisease AT kwangsiknho deepfusionofincompletemultiomicdataformolecularmechanismofalzheimersdisease AT paulsalama deepfusionofincompletemultiomicdataformolecularmechanismofalzheimersdisease AT andrewjsaykin deepfusionofincompletemultiomicdataformolecularmechanismofalzheimersdisease AT shiaofenfang deepfusionofincompletemultiomicdataformolecularmechanismofalzheimersdisease AT jingwenyan deepfusionofincompletemultiomicdataformolecularmechanismofalzheimersdisease |