Short-read RNA-seq yields lower estimates of A-to-I RNA editing levels than long-read cDNA sequencing

Area of Science:

• Molecular Biology
• Genomics
• Bioinformatics

Background:

• Adenosine to inosine (A-to-I) RNA editing is a frequent RNA modification crucial for RNA metabolism.
• Accurate quantification of A-to-I editing levels is vital, yet often overlooked, especially when comparing Next-Generation Sequencing (NGS) and Long-Read Sequencing (LRS) RNA-seq.

Purpose of the Study:

• To investigate and compare the accuracy of A-to-I RNA editing quantification between NGS and LRS RNA-seq.
• To identify factors influencing discrepancies in editing level estimations.

Main Methods:

• Paired NGS and LRS cDNA RNA-sequencing of HEK293T and U2OS cells.
• Full-length amplicon sequencing for validation.
• Analysis of public NGS and LRS RNA-seq datasets from various human cell lines, including cancer cell lines.

Main Results:

• NGS consistently estimated lower A-to-I editing levels compared to LRS.
• This underestimation was confirmed across multiple cell lines and datasets.
• Sequencing read length was identified as a primary factor, with shorter reads showing a positive correlation with lower quantification accuracy.
• Shorter reads exhibited increased alignment errors, contributing to the underestimation of editing levels.

Conclusions:

• Long-Read Sequencing (LRS) RNA-seq provides more accurate quantification of A-to-I RNA editing levels than Next-Generation Sequencing (NGS).
• Read length significantly impacts the accuracy of RNA editing quantification.
• LRS RNA-seq is recommended for precise assessment of A-to-I editing levels.