RIVER

RIVER estimates the regulatory impact of rare genetic variants on gene expression by integrating individual whole genome sequencing (WGS) and RNA-sequencing (RNA-seq) data.

Key Features:

• Probabilistic modeling framework: Integrates whole genome sequencing (WGS) and RNA-sequencing (RNA-seq) data from the same individuals for joint genomic and transcriptomic analysis.
• Bayesian approach: Uses Bayesian statistics to estimate the probability that a given genetic variant has a regulatory impact on gene expression within an individual.
• Integration with GTEx data: Leverages Genotype-Tissue Expression (GTEx) whole-genome and multi-tissue RNA-sequencing datasets to analyze expression across tissues.
• Identification of functional variants: Associates rare variants with gene expression outliers, reporting that 58% of underexpression and 28% of overexpression outliers have underlying rare variants versus 9% in non-outliers.
• Focus on conserved and proximal regulatory sites: Emphasizes variants near transcription start sites (TSS) and at evolutionarily conserved regions as enriched for large expression effects.
• Application to disease genes: Identifies known disease genes with expression outliers to aid prioritization of functional regulatory variants impacting disease risk.

Scientific Applications:

• Personal genome interpretation: Assess the regulatory impact of rare variants in individual genomes by combining WGS and RNA-seq data.
• Genetic epidemiology: Improve attribution of expression changes to rare regulatory variants in population-scale datasets.
• Prioritization of functional regulatory variants: Rank rare variants near TSS or conserved sites for downstream functional or clinical follow-up.
• Study of disease mechanisms: Identify expression outliers in known disease genes to link rare variants to potential pathogenic mechanisms.
• Multi-tissue expression analysis: Analyze gene expression outliers across multiple tissues using GTEx multi-tissue RNA-seq data.

Methodology:

Applies a probabilistic modeling framework with Bayesian statistical estimation to jointly analyze WGS and RNA-seq from the same individuals, uses GTEx whole-genome and multi-tissue RNA-sequencing data to identify gene expression outliers across tissues, and associates rare variants—including those near TSS and at conserved sites—with expression changes.