Abstract
Motivation In the past few years many prediction approaches have been proposed and widely employed in high dimensional genetic data for disease risk evaluation. However, those approaches typically ignore in model fitting the important group structures that naturally exists in genetic data. Methods In the present study, we applied a novel model-averaging approach, called jackknife model averaging prediction (JMAP), for high dimensional genetic risk prediction while incorporating pathway information into the model specification. JMAP selects the optimal weights across candidate models by minimizing a cross validation criterion in a jackknife way. Compared with previous approaches, one of the primary features of JMAP is to allow model weights to vary from 0 to 1 but without the limitation that the summation of weights is equal to one. We evaluated the performance of JMAP using extensive simulation studies and compared it with existing methods. We finally applied JMAP to four real cancer datasets that are publicly available from TCGA. Results The simulations showed that compared with other existing approaches (e.g., gsslasso), JMAP performed best or is among the best methods across a range of scenarios. For example, among 14 out of 16 simulation settings with PVE = 0.3, JMAP has an average of 0.075 higher prediction accuracy compared with gsslasso. We further found that in the simulation, the model weights for the true candidate models have much smaller chances to be zero compared with those for the null candidate models and are substantially greater in magnitude. In the real data application, JMAP also behaves comparably or better compared with the other methods for continuous phenotypes. For example, for the COAD, CRC, and PAAD datasets, the average gains of predictive accuracy of JMAP are 0.019, 0.064, and 0.052 compared with gsslasso. Conclusion The proposed method JMAP is a novel model-averaging approach for high dimensional genetic risk prediction while incorporating external useful group structures into the model specification.
Highlights
jackknife model averaging prediction (JMAP) consists of two-step model fitting procedures: (i) in the first step, we divide the molecular predictors into K biological pathways/groups (e.g., Kyoto Encyclopedia of Genes and Genomes (KEGG)) and build a series of candidate linear prediction models with gene expression measurements available for various groups; we assume that the pathways are predetermined and that the predictors may overlap across different pathways; (ii) in the second step, we look for a suitable weight vector for averaging across the candidate models to perform a pooled prediction
In the setting with 300 group in scenario III, all the four competitive methods (i.e., Lasso, ENET, random forest, and gsslasso) have a higher prediction accuracy relative to JMAP. e simulation results for phenotypic variance explained (PVE) 0.5 and 0.8 are displayed in Figures S2–S5 in Supplementary Materials; we observed the similar pattern that JMAP performs better or is as good as other competing methods in most of the simulated settings
In the PAAD dataset, JMAP is better than Lasso, gsslasso, and ENET, while random forest has the highest prediction accuracy
Summary
Due to the rapid development of biotechnology [1,2,3,4], a large number of high-throughput and low-cost genetic datasets have been generated and provide a broad space to investigate the association between genetic markers and complex diseases/disorders [5,6,7,8,9,10,11,12,13,14]. e great success of association studies further promotes the risk prediction and evaluation for complex phenotypes by incorporating into genetic information (e.g., gene expressions or single nucleotide polymorphisms) [15,16,17,18,19,20]. In the past few years, developing prediction methods that can efficiently model high dimensional genetic data has been an active area and attracted much research attention, and a Computational and Mathematical Methods in Medicine series of novel prediction approaches have been proposed and widely employed for disease risk evaluation or gene expression imputation [21,22,23,24,25,26,27]. Most of those approaches ignore in model fitting the important information of group structures or functional classifications that naturally exist in genetic data. One of the widely-used group sources is the pathway information in the Kyoto Encyclopedia of Genes and Genomes (KEGG) [37, 38], which integrates information on genomic, chemical, and system functions and groups genes with highly related sequences in terms of the sequence similarity of genes
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Computational and Mathematical Methods in Medicine
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
