Allele Specific Expression Quality Control Fills Critical Gap in Transcriptome Assisted Rare Variant Interpretation.

Area of Science:

• Genomics
• Transcriptomics
• Bioinformatics

Background:

• Allele-specific expression (ASE) reveals genetic variation's impact on transcription but is susceptible to artifacts.
• High-quality data is crucial for accurate cis-regulatory effect analysis.

Purpose of the Study:

• Introduce aseQC, a statistical framework to quantify sample-level ASE quality.
• Improve the robustness of downstream transcriptome analyses by excluding noisy samples.

Main Methods:

• Developed a statistical framework (aseQC) to measure extra-binomial variation in ASE data.
• Applied aseQC to datasets including rare Mendelian muscular disorders and GTEx project data.
• Evaluated the impact of low-quality samples on rare and common variant analyses.

Main Results:

• aseQC identified low-quality samples in rare disease cohorts, demonstrating clinical utility.
• Discovered 563 low-quality samples in GTEx data, linked to processing dates and exhibiting excessive allelic imbalance.
• Low-quality samples significantly increased ASE and splicing outliers, degrading rare variant interpretation but not affecting common variant analysis.

Conclusions:

• aseQC provides a reliable method for assessing sample quality in transcriptome data.
• Excluding low-quality samples is critical for accurate rare variant analysis.
• aseQC enhances the reliability of genomic studies utilizing allele-specific expression data.