WaveQTL

WaveQTL applies wavelet-based statistical methods to detect genetic associations with high-resolution functional phenotypes derived from high-throughput sequencing assays.

Key Features:

• Wavelet-based functional data analysis: Treats sequence-derived measurements as continuous functions varying along the genome using wavelet-based statistical methods.
• High-resolution data utilization: Operates on assay data from RNA-seq, ChIP-seq, and DNase-seq to capture fine-scale molecular and cellular phenotypic variation.
• Flexible regional modeling: Implements wavelet regression for regional tests that model genetic effects at multiple scales compared to conventional GWAS windowing.
• Zooming strategy: Applies a zooming approach to enhance detection of associations in small genomic regions.
• Improved performance in high LD: Includes modifications that increase power in regions with high linkage disequilibrium between SNPs.

Scientific Applications:

• Molecular phenotype association mapping: Mapping genetic variants to gene expression, transcription factor binding, and chromatin accessibility at high resolution.
• Detection of dsQTLs: Identifying DNA sequence quantitative trait loci associated with chromatin accessibility.
• Regional association discovery: Detecting loci that may be missed by conventional GWAS through multiscale regional testing.
• Replication and fine-mapping of complex-trait loci: Replicating and localizing loci for traits such as birth weight, including in smaller sample sizes.

Methodology:

Uses wavelet-based statistical methods and wavelet regression to model sequence-derived functional phenotypes, incorporates a zooming strategy for fine-scale regional testing, and is applied to high-resolution sequencing assays (RNA-seq, ChIP-seq, DNase-seq) with adjustments to improve power in high-LD regions.