HybridMTest

HybridMTest performs hybrid multiple testing to analyze high-throughput genomic and molecular biology data by integrating method selection and assumption evaluations with empirical Bayes probability (EBP) estimates derived from Grenander density estimation.

Key Features:

• Hybrid Multiple Testing: Performs multiple testing that combines various statistical methods to optimize analysis accuracy.
• Method Selection: Incorporates selection of statistical methods appropriate to specific datasets.
• Assumption Evaluations: Evaluates underlying data assumptions to support robust inference.
• Empirical Bayes Probability (EBP) Estimates: Derives EBP estimates using Grenander density estimation to provide a probabilistic framework for decision making.
• Bioconductor Integration: Distributed as part of the Bioconductor project for interoperability with other Bioconductor packages.

Scientific Applications:

• High-throughput genomics: Analysis of high-throughput genomic data where multiple testing corrections and EBP estimates are required.
• Molecular biology data analysis: Application to molecular biology datasets that benefit from method selection and assumption evaluation to improve result reliability.

Methodology:

Applies Grenander density estimation to derive empirical Bayes probability (EBP) estimates, implements hybrid multiple testing by combining various statistical methods, and incorporates explicit method selection and evaluation of underlying assumptions.