Efficient and comprehensive representation of uniqueness for next-generation sequencing by minimum unique length...

Area of Science:

• Genomics
• Bioinformatics
• Transcriptomics

Background:

• Next-generation sequencing and RNA-Seq are increasingly used for transcriptome studies.
• Computational analysis of RNA-Seq data involves mapping short reads to the genome, often resulting in multimapping reads.
• Multimapping reads pose challenges for accurate transcript expression quantification.

Purpose of the Study:

• To develop a framework (MULTo) for representing mappability information by identifying the minimum unique length for genomic coordinates.
• To compare uniqueness compensation approaches for transcript expression quantification using MULTo.
• To facilitate the use of uniqueness compensation in RNA-Seq analysis and reduce reliance on external mappability files.

Main Methods:

• Developed the Minimum Unique Length Tool (MULTo) to calculate the shortest unique length for each genomic coordinate.
• Utilized minimum unique length information to compare different strategies for handling multimapping reads in transcript expression quantification.
• Explored uniqueness in specific mouse genomic regions and enhancer mapping.

Main Results:

• Demonstrated that discarding multimapping reads and adjusting gene model lengths provides the best compensation for transcript expression quantification.
• MULTo efficiently represents mappability information.
• Identified specific regions with varying uniqueness in the mouse genome.

Conclusions:

• The Minimum Unique Length Tool (MULTo) provides a robust method for assessing genomic uniqueness in RNA-Seq data.
• Discarding multimapping reads and adjusting gene models is the optimal strategy for accurate transcript quantification.
• MULTo aims to standardize and improve RNA-Seq analysis by integrating mappability information directly.