Enhancing transcriptome expression quantification through accurate assignment of long RNA sequencing reads with...

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• Long-read RNA sequencing offers comprehensive transcriptomes but faces challenges due to higher error rates.
• Existing computational tools for long-read data show inconsistencies in transcript assembly and quantification.
• Accurate read assignment is crucial for reliable transcriptome characterization.

Purpose of the Study:

• To introduce TranSigner, a novel computational tool for assigning long RNA sequencing reads to transcriptomes.
• To enhance the accuracy of transcript abundance estimation using long-read data.
• To provide researchers with greater confidence in transcriptome analysis.

Main Methods:

• TranSigner employs three modules: read alignment, compatibility computation, and expectation-maximization for probabilistic assignment.
• The tool aligns long reads to a provided transcriptome reference.
• It calculates read-to-transcript compatibility using alignment scores and positions.

Main Results:

• TranSigner demonstrates accurate read assignment capabilities on simulated and experimental datasets.
• The tool achieves higher accuracy in transcript abundance estimation compared to existing methods.
• Validation was performed on data from Homo sapiens, Arabidopsis thaliana, and Mus musculus.

Conclusions:

• TranSigner effectively addresses uncertainties in long-read transcriptome analysis.
• The tool improves the reliability of transcript abundance quantification.
• TranSigner is a versatile solution for characterizing transcript support at the read level.