Identification and quantification of small exon-containing isoforms in long-read RNA sequencing data

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Small exons, including microexons (≤30 nucleotides), are common in gene transcriptomes and vital for gene function.
• Long-read RNA sequencing (RNA-seq) using Oxford Nanopore Technologies (ONT) provides full transcript coverage but faces challenges in accurately quantifying small exons due to lower base accuracy.
• Misaligned reads containing small exons can lead to inaccurate transcript quantification.

Purpose of the Study:

• To develop and validate a computational method for improving the quantification of small exons in ONT long-read RNA-seq data.
• To assess the impact of misaligned reads on small exon quantification.
• To identify novel neural-specific microexons and understand their regulatory roles.

Main Methods:

• Systematic assessment of small exon quantification using synthetic and human ONT RNA-seq datasets.
• Development of MisER (Misaligned Exons Remapper), a local-realignment method to correct misaligned reads containing small exons.
• Validation of MisER's sensitivity and specificity using synthetic and simulated datasets.
• Comparison of microexon Percent Spliced-In (PSI) index across 14 neural and 16 non-neural human tissues.

Main Results:

• Reads containing small exons are frequently misaligned in standard ONT RNA-seq analysis, compromising transcript quantification.
• MisER effectively remapped misaligned reads, demonstrating high sensitivity and specificity for quantifying transcripts with small exons.
• MisER identified a higher Percent Spliced-In (PSI) index for microexons in neural tissues compared to non-neural tissues, highlighting neural-specific microexon regulation.
• The method facilitates the discovery of functionally relevant small exons.

Conclusions:

• MisER significantly enhances the accuracy of small exon quantification from ONT long-read RNA-seq data.
• This advancement is crucial for studying gene regulation involving small and microexons, particularly in neurobiology.
• The findings provide a valuable tool for researchers investigating transcript isoform variation and its functional implications, especially in neural development and function.