Unique Molecular Identifiers reveal a novel sequencing artefact with implications for RNA-Seq based gene expression...

Area of Science:

• Molecular Biology
• Genomics
• Bioinformatics

Background:

• Unique Molecular Identifiers (UMI) are crucial for accurate RNA sequencing by tagging individual RNA molecules.
• PCR amplification bias is a significant challenge in RNA-Seq, especially in single-cell studies requiring extensive amplification.

Purpose of the Study:

• To investigate the phenomenon of reads with identical UMIs mapping to disparate genomic locations.
• To identify the underlying causes of this mapping discrepancy and its implications for transcript abundance estimation.

Main Methods:

• Analysis of single-cell RNA-Sequencing (scRNA-Seq) data from Physcomitrella patens.
• Examination of publicly available UMI-containing RNA-Seq datasets.
• Investigation of PCR stutter as a potential explanation for mapping anomalies.

Main Results:

• Reads sharing a UMI, presumed to originate from the same RNA molecule, were frequently observed to map to different genomic coordinates.
• This artifact was consistent across multiple libraries and datasets, suggesting a widespread issue.
• PCR stutter, particularly in low-entropy or repetitive sequences, was identified as a likely contributor to the observed mapping discrepancies.

Conclusions:

• The assumption that reads with different mapping coordinates originate from distinct molecules is not always valid when using UMIs.
• Failure to account for UMI-masked PCR stutter can lead to an overestimation of transcript abundances.
• This finding necessitates re-evaluation of UMI-based quantification strategies in RNA sequencing.