A junction coverage compatibility score to quantify the reliability of transcript abundance estimates and annotation...

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• RNA sequencing (RNA-seq) data analysis relies heavily on accurate abundance estimates of genomic features.
• The quality of downstream results is directly contingent upon the reliability of these abundance estimations.
• Existing methods often assume accurate transcript abundance and annotation data.

Purpose of the Study:

• To introduce a novel metric, the junction coverage compatibility score, for evaluating transcript-level abundance estimates.
• To assess the accuracy of transcript annotation catalogs using RNA-seq data.
• To identify potential sources of error in RNA-seq data analysis pipelines.

Main Methods:

• Developed the junction coverage compatibility score by comparing observed and predicted splice junction read counts.
• Inferred predicted junction-spanning reads from estimated transcript abundances and genomic coordinates.
• Analyzed RNA-seq data to identify genes with discrepancies between observed and predicted junction coverages.

Main Results:

• The junction coverage compatibility score effectively evaluates the reliability of transcript abundance estimates.
• A small subset of genes consistently shows poor agreement between observed and predicted junction coverages.
• This poor agreement is independent of the specific transcript abundance estimation method used.

Conclusions:

• Transcript abundance estimates for genes with poor junction coverage compatibility should be interpreted with caution.
• The junction coverage compatibility score is a valuable tool for quality control in RNA-seq data analysis.
• This metric aids in identifying and addressing potential inaccuracies in transcript annotation and abundance estimation.