Integrated Analysis of Pharmacokinetic, Clinical, and SNP Microarray Data Using Projection onto the Most Interesting Statistical Evidence with Adaptive Permutation Testing

Powerful methods for integrated analysis of multiple biological data sets must be developed to maximize researchers' ability to interpret them and acquire meaningful knowledge. Projection onto the most interesting statistical evidence (PROMISE) is a powerful statistical procedure that has been recently developed to incorporate a biological paradigm for the relationships among endpoint variables into an integrated analysis of microarray gene expression data with multiple biological and clinical endpoints. Here, PROMISE is adapted to perform an integrated analysis of pharmacokinetic, clinical, and genome-wide genotype data that incorporates a clinically relevant biological paradigm for pharmacokinetic and clinical response data. An efficient permutation-testing algorithm is introduced so that statistical calculations are computationally feasible in this higher-dimension setting. The new method is applied to a pediatric leukemia data set. The results clearly indicate that PROMISE can be a very powerful statistical tool for identifying genomic features that exhibit a pattern of association with multiple endpoint variables which is concordant with a practically useful biological paradigm.