geQTL

geQTL applies a sparse regression framework to detect individual and group-wise associations between single-nucleotide polymorphisms (SNPs) and gene expression for expression quantitative trait loci (eQTL) analysis.

Key Features:

• Sparse regression method: Implements a novel sparse regression method to identify associations between SNPs and expression traits.
• Individual and group-wise detection: Detects both single SNP–gene associations and collective SNP effects on groups of genes.
• Confounder correction: Corrects for potential confounders to reduce spurious associations.
• Computational efficiency: Employs computationally efficient techniques suitable for large-scale studies.
• Automatic group inference: Automatically infers the appropriate number of group-wise associations.
• Decoupling of signals: Decouples individual from group-wise associations to separate single-variant and multi-variant signals.
• Empirical validation: Validated on simulated and yeast datasets and reported to outperform state-of-the-art methods.

Scientific Applications:

• eQTL mapping: Mapping expression quantitative trait loci to study genetic regulation of gene expression in complex traits and common diseases.
• Pathway-level analysis: Detecting pathway- or module-level genetic influences by modeling group-wise SNP effects.
• Gene expression network analysis: Exploring genetic influences on gene expression networks by separating individual and collective SNP effects.
• Method benchmarking: Benchmarking and validation of eQTL methods using simulated and yeast datasets.

Methodology:

Uses a sparse regression algorithm to detect individual and group-wise SNP–gene associations, includes confounder correction, applies computationally efficient techniques, automatically infers the number of group-wise associations, and decouples individual from group-wise signals; validated on simulated and yeast datasets.