Semiparametric tests for identifying differentially methylated loci with case-control designs using Illumina arrays.

摘要: DNA methylation plays an important role in the development of many types cancer. Identifying differentially methylated loci between cancer and normal patients is one central tasks to understand contributions process on development. Through investigation measurements generated by Illumina arrays, we notice that groups could differ not only means but also variances. Therefore, propose a generalized exponential tilt model capture differences both variances groups. We derive semiparametric tests obtain robustness. simulation studies, demonstrate feasibility proposed much improved power than t-test regression-based when are different or Hence can serve as useful complements standard test means. illustrate methods applying real data from recent study ovarian where identified additional were missed existing method.