Selective transcriptional profiling is an attractive approach for alleviating the high cost of genetical genomics research as it requires only a subset of individuals in the QTL mapping study for microarray experiments. Current statistical methods for this approach are based on parametric models that might not be appropriate for all experiments. To provide a nonparametric method for analyzing data obtained in selective transcriptional profiling studies, an empirical-likelihood-based inference is derived for multi-sample comparisons when information is available on surrogate variables. The results show that when testing for the association between the transcriptional abundance of a given gene and a known QTL, using relatively inexpensive trait data on extra individuals significantly improves the power for the proposed test.
ASJC Scopus subject areas
- Statistics and Probability
- Computational Mathematics
- Computational Theory and Mathematics
- Applied Mathematics