Coherent Multi-Task Feature Selection and Prediction from Pharmacogenomics Databases (304830)*Alahendra Acharige Chamila Dilhani Perera, Texas Tech University
Raziur Rahman, Texas Tech University
Ranadip Pal, Texas Tech University
Souparno Ghosh, Texas Tech University
Keywords: Adaptive multi-task elastic net, Random Forest, Bayesian data fusion model, Pharmacogenomics data
Integrating multiple databases of similar tasks is a significant problem in biological data analysis. In this article we offer two methodologies for combining information across similar databases. First, we demonstrate that an adaptive multi-task elastic net for feature selection and Random Forest for prediction, can be used to borrow information across databases with superior predictive performance. Second, we offer a two stage Bayesian data fusion model to perform full predictive inference. We illustrate both methodologies with synthetic data and publicly available Pharmacogenomics data.