Transcriptome-wide root causal inference.

Area of Science:

• Genetics
• Computational Biology
• Systems Biology

Background:

• Root causal genes initiate disease by being the first to change expression.
• Targeting these genes could halt disease progression.
• Current methods cannot identify root causal genes using only observational data.

Purpose of the Study:

• To introduce the Transcriptome-Wide Root Causal Inference (TWRCI) algorithm.
• To identify root causal genes and their causal network from genetic variants and RNA sequencing data.
• To address the limitation of existing algorithms in discovering root causal genes from observational data alone.

Main Methods:

• Developed the Transcriptome-Wide Root Causal Inference (TWRCI) algorithm.
• Utilized a competitive regression method to link genetic variants to gene expression.
• Simultaneously determined gene expression propagation sequence and causal graph.
• Integrated genetic variant data with unperturbed bulk RNA sequencing data.

Main Results:

• TWRCI successfully identifies root causal genes and their causal graph.
• The algorithm outperforms existing methods on multiple metrics.
• Demonstrated TWRCI's efficacy by uncovering root causal mechanisms in two complex diseases.
• Confirmed findings through replication with independent genome-wide summary statistics.

Conclusions:

• TWRCI is a novel algorithm for discovering root causal genes from observational data.
• The algorithm accounts for complex genetic factors like pleiotropy and heterogeneity.
• This method offers a new avenue for understanding and potentially treating complex diseases at their origin.