FFW

FFW detects differentially methylated regions (DMRs) in DNA methylation data using a wavelet-based analysis to identify associations between functional genomic signals and phenotypic traits, including QTLs associated with DNase I hypersensitivity.

Key Features:

• Wavelet-Based Analysis: Applies wavelet transformation to DNA methylation data to capture localized changes in methylation and identify quantitative trait loci (QTLs) linked to DNase I hypersensitivity.
• Simulation-Based Significance Testing: Estimates significance by calculating the distribution of Bayes factors via simulation rather than permutation-based methods.
• Computational Efficiency: Incorporates methodological innovations from Zhou and Guan to substantially increase computational speed relative to the WaveQTL framework.
• Type I Error Control and Power: Demonstrates control of type I error rates while maintaining power for DMR detection in analyses of publicly available DNA methylation datasets, including colorectal cancer data.
• Broad Applicability: Extends to detecting associations between diverse types of functional genomic data and phenotypic traits beyond DNA methylation.

Scientific Applications:

• Epigenetic DMR discovery: Identification of DMRs in DNA methylation datasets to investigate disease mechanisms such as colorectal cancer.
• Functional QTL mapping: Detection of QTLs associated with DNase I hypersensitivity and other functional genomic signals.
• Re-analysis of public datasets: Re-analysis of existing methylation datasets from public repositories to uncover previously overlooked associations relevant to genetic regulation and disease etiology.

Methodology:

Performs wavelet transformation of methylation signals; computes Bayes factors and estimates their significance via simulation-derived null distributions instead of permutation; builds on the WaveQTL framework and incorporates computational optimizations from Zhou and Guan.