miSTAR : miRNA target prediction through modeling quantitative and qualitative miRNA binding site information in a stacked model structure
- Author
- Gert Van Peer (UGent) , Ayla De Paepe, Michiel Stock (UGent) , Jasper Anckaert (UGent) , Pieter-Jan Volders (UGent) , Jo Vandesompele (UGent) , Bernard De Baets (UGent) and Willem Waegeman (UGent)
- Organization
- Abstract
- In microRNA (miRNA) target prediction, typically two levels of information need to be modeled: the number of potential miRNA binding sites present in a target mRNA and the genomic context of each individual site. Single model structures insufficiently cope with this complex training data structure, consisting of feature vectors of unequal length as a consequence of the varying number of miRNA binding sites in different mRNAs. To circumvent this problem, we developed a two-layered, stacked model, in which the influence of binding site context is separately modeled. Using logistic regression and random forests, we applied the stacked model approach to a unique data set of 7990 probed miRNA-mRNA interactions, hereby including the largest number of miRNAs in model training to date. Compared to lower-complexity models, a particular stacked model, named miSTAR (miRNA stacked model target prediction; www.mi-star.org), displays a higher general performance and precision on top scoring predictions. More importantly, our model outperforms published and widely used miRNA target prediction algorithms. Finally, we highlight flaws in cross-validation schemes for evaluation of miRNA target prediction models and adopt a more fair and stringent approach.
- Keywords
- MICRORNA TARGETS, IDENTIFICATION, REPRESSION, CURVES
Downloads
-
gkw1260.pdf
- full text
- |
- open access
- |
- |
- 2.76 MB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU-8509343
- MLA
- Van Peer, Gert, et al. “MiSTAR : MiRNA Target Prediction through Modeling Quantitative and Qualitative MiRNA Binding Site Information in a Stacked Model Structure.” NUCLEIC ACIDS RESEARCH, vol. 45, no. 7, 2017, doi:10.1093/nar/gkw1260.
- APA
- Van Peer, G., De Paepe, A., Stock, M., Anckaert, J., Volders, P.-J., Vandesompele, J., … Waegeman, W. (2017). miSTAR : miRNA target prediction through modeling quantitative and qualitative miRNA binding site information in a stacked model structure. NUCLEIC ACIDS RESEARCH, 45(7). https://doi.org/10.1093/nar/gkw1260
- Chicago author-date
- Van Peer, Gert, Ayla De Paepe, Michiel Stock, Jasper Anckaert, Pieter-Jan Volders, Jo Vandesompele, Bernard De Baets, and Willem Waegeman. 2017. “MiSTAR : MiRNA Target Prediction through Modeling Quantitative and Qualitative MiRNA Binding Site Information in a Stacked Model Structure.” NUCLEIC ACIDS RESEARCH 45 (7). https://doi.org/10.1093/nar/gkw1260.
- Chicago author-date (all authors)
- Van Peer, Gert, Ayla De Paepe, Michiel Stock, Jasper Anckaert, Pieter-Jan Volders, Jo Vandesompele, Bernard De Baets, and Willem Waegeman. 2017. “MiSTAR : MiRNA Target Prediction through Modeling Quantitative and Qualitative MiRNA Binding Site Information in a Stacked Model Structure.” NUCLEIC ACIDS RESEARCH 45 (7). doi:10.1093/nar/gkw1260.
- Vancouver
- 1.Van Peer G, De Paepe A, Stock M, Anckaert J, Volders P-J, Vandesompele J, et al. miSTAR : miRNA target prediction through modeling quantitative and qualitative miRNA binding site information in a stacked model structure. NUCLEIC ACIDS RESEARCH. 2017;45(7).
- IEEE
- [1]G. Van Peer et al., “miSTAR : miRNA target prediction through modeling quantitative and qualitative miRNA binding site information in a stacked model structure,” NUCLEIC ACIDS RESEARCH, vol. 45, no. 7, 2017.
@article{8509343,
abstract = {{In microRNA (miRNA) target prediction, typically two levels of information need to be modeled: the number of potential miRNA binding sites present in a target mRNA and the genomic context of each individual site. Single model structures insufficiently cope with this complex training data structure, consisting of feature vectors of unequal length as a consequence of the varying number of miRNA binding sites in different mRNAs. To circumvent this problem, we developed a two-layered, stacked model, in which the influence of binding site context is separately modeled. Using logistic regression and random forests, we applied the stacked model approach to a unique data set of 7990 probed miRNA-mRNA interactions, hereby including the largest number of miRNAs in model training to date. Compared to lower-complexity models, a particular stacked model, named miSTAR (miRNA stacked model target prediction; www.mi-star.org), displays a higher general performance and precision on top scoring predictions. More importantly, our model outperforms published and widely used miRNA target prediction algorithms. Finally, we highlight flaws in cross-validation schemes for evaluation of miRNA target prediction models and adopt a more fair and stringent approach.}},
articleno = {{e51}},
author = {{Van Peer, Gert and De Paepe, Ayla and Stock, Michiel and Anckaert, Jasper and Volders, Pieter-Jan and Vandesompele, Jo and De Baets, Bernard and Waegeman, Willem}},
issn = {{0305-1048}},
journal = {{NUCLEIC ACIDS RESEARCH}},
keywords = {{MICRORNA TARGETS,IDENTIFICATION,REPRESSION,CURVES}},
language = {{eng}},
number = {{7}},
pages = {{9}},
title = {{miSTAR : miRNA target prediction through modeling quantitative and qualitative miRNA binding site information in a stacked model structure}},
url = {{http://doi.org/10.1093/nar/gkw1260}},
volume = {{45}},
year = {{2017}},
}
- Altmetric
- View in Altmetric
- Web of Science
- Times cited: