Min-redundancy and max-relevance multi-view feature selection for predicting ovarian cancer survival using multi-omics data
Abstract Background Large-scale collaborative precision medicine initiatives (e.g., The Cancer Genome Atlas (TCGA)) are yielding rich multi-omics data. Integrative analyses of the resulting multi-omics data, such as somatic mutation, copy number alteration (CNA), DNA methylation, miRNA, gene express...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2018-09-01
|
| Series: | BMC Medical Genomics |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1186/s12920-018-0388-0 |
| _version_ | 1818728259477569536 |
|---|---|
| author | Yasser EL-Manzalawy Tsung-Yu Hsieh Manu Shivakumar Dokyoon Kim Vasant Honavar |
| author_facet | Yasser EL-Manzalawy Tsung-Yu Hsieh Manu Shivakumar Dokyoon Kim Vasant Honavar |
| author_sort | Yasser EL-Manzalawy |
| collection | DOAJ |
| description | Abstract Background Large-scale collaborative precision medicine initiatives (e.g., The Cancer Genome Atlas (TCGA)) are yielding rich multi-omics data. Integrative analyses of the resulting multi-omics data, such as somatic mutation, copy number alteration (CNA), DNA methylation, miRNA, gene expression, and protein expression, offer tantalizing possibilities for realizing the promise and potential of precision medicine in cancer prevention, diagnosis, and treatment by substantially improving our understanding of underlying mechanisms as well as the discovery of novel biomarkers for different types of cancers. However, such analyses present a number of challenges, including heterogeneity, and high-dimensionality of omics data. Methods We propose a novel framework for multi-omics data integration using multi-view feature selection. We introduce a novel multi-view feature selection algorithm, MRMR-mv, an adaptation of the well-known Min-Redundancy and Maximum-Relevance (MRMR) single-view feature selection algorithm to the multi-view setting. Results We report results of experiments using an ovarian cancer multi-omics dataset derived from the TCGA database on the task of predicting ovarian cancer survival. Our results suggest that multi-view models outperform both view-specific models (i.e., models trained and tested using a single type of omics data) and models based on two baseline data fusion methods. Conclusions Our results demonstrate the potential of multi-view feature selection in integrative analyses and predictive modeling from multi-omics data. |
| first_indexed | 2024-12-17T22:27:09Z |
| format | Article |
| id | doaj.art-28bc795645b940629925263fac0cda00 |
| institution | Directory Open Access Journal |
| issn | 1755-8794 |
| language | English |
| last_indexed | 2024-12-17T22:27:09Z |
| publishDate | 2018-09-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Genomics |
| spelling | doaj.art-28bc795645b940629925263fac0cda002022-12-21T21:30:18ZengBMCBMC Medical Genomics1755-87942018-09-0111S3193110.1186/s12920-018-0388-0Min-redundancy and max-relevance multi-view feature selection for predicting ovarian cancer survival using multi-omics dataYasser EL-Manzalawy0Tsung-Yu Hsieh1Manu Shivakumar2Dokyoon Kim3Vasant Honavar4Artificial Intelligence Research Laboratory, College of Information Sciences and Technology, Pennsylvania State UniversityArtificial Intelligence Research Laboratory, College of Information Sciences and Technology, Pennsylvania State UniversityBiomedical and Translational Informatics Institute, Geisinger Health SystemBiomedical and Translational Informatics Institute, Geisinger Health SystemArtificial Intelligence Research Laboratory, College of Information Sciences and Technology, Pennsylvania State UniversityAbstract Background Large-scale collaborative precision medicine initiatives (e.g., The Cancer Genome Atlas (TCGA)) are yielding rich multi-omics data. Integrative analyses of the resulting multi-omics data, such as somatic mutation, copy number alteration (CNA), DNA methylation, miRNA, gene expression, and protein expression, offer tantalizing possibilities for realizing the promise and potential of precision medicine in cancer prevention, diagnosis, and treatment by substantially improving our understanding of underlying mechanisms as well as the discovery of novel biomarkers for different types of cancers. However, such analyses present a number of challenges, including heterogeneity, and high-dimensionality of omics data. Methods We propose a novel framework for multi-omics data integration using multi-view feature selection. We introduce a novel multi-view feature selection algorithm, MRMR-mv, an adaptation of the well-known Min-Redundancy and Maximum-Relevance (MRMR) single-view feature selection algorithm to the multi-view setting. Results We report results of experiments using an ovarian cancer multi-omics dataset derived from the TCGA database on the task of predicting ovarian cancer survival. Our results suggest that multi-view models outperform both view-specific models (i.e., models trained and tested using a single type of omics data) and models based on two baseline data fusion methods. Conclusions Our results demonstrate the potential of multi-view feature selection in integrative analyses and predictive modeling from multi-omics data.http://link.springer.com/article/10.1186/s12920-018-0388-0Multi-omics data integrationMulti-view feature selectionCancer survival predictionMachine learning |
| spellingShingle | Yasser EL-Manzalawy Tsung-Yu Hsieh Manu Shivakumar Dokyoon Kim Vasant Honavar Min-redundancy and max-relevance multi-view feature selection for predicting ovarian cancer survival using multi-omics data BMC Medical Genomics Multi-omics data integration Multi-view feature selection Cancer survival prediction Machine learning |
| title | Min-redundancy and max-relevance multi-view feature selection for predicting ovarian cancer survival using multi-omics data |
| title_full | Min-redundancy and max-relevance multi-view feature selection for predicting ovarian cancer survival using multi-omics data |
| title_fullStr | Min-redundancy and max-relevance multi-view feature selection for predicting ovarian cancer survival using multi-omics data |
| title_full_unstemmed | Min-redundancy and max-relevance multi-view feature selection for predicting ovarian cancer survival using multi-omics data |
| title_short | Min-redundancy and max-relevance multi-view feature selection for predicting ovarian cancer survival using multi-omics data |
| title_sort | min redundancy and max relevance multi view feature selection for predicting ovarian cancer survival using multi omics data |
| topic | Multi-omics data integration Multi-view feature selection Cancer survival prediction Machine learning |
| url | http://link.springer.com/article/10.1186/s12920-018-0388-0 |
| work_keys_str_mv | AT yasserelmanzalawy minredundancyandmaxrelevancemultiviewfeatureselectionforpredictingovariancancersurvivalusingmultiomicsdata AT tsungyuhsieh minredundancyandmaxrelevancemultiviewfeatureselectionforpredictingovariancancersurvivalusingmultiomicsdata AT manushivakumar minredundancyandmaxrelevancemultiviewfeatureselectionforpredictingovariancancersurvivalusingmultiomicsdata AT dokyoonkim minredundancyandmaxrelevancemultiviewfeatureselectionforpredictingovariancancersurvivalusingmultiomicsdata AT vasanthonavar minredundancyandmaxrelevancemultiviewfeatureselectionforpredictingovariancancersurvivalusingmultiomicsdata |