Ryūtō: improved multi-sample transcript assembly for differential transcript expression analysis and more.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Accurate RNA-sequencing (RNA-seq) assembly is vital for gene annotation and expression studies.
• Traditional single-sample assembly methods have limitations.
• Multi-sample RNA-seq experiments offer more data but face challenges with error accumulation.

Purpose of the Study:

• To present an extension of the Ryūtō tool for reconstructing consensus transcriptomes from multiple RNA-seq datasets.
• To improve the accuracy and efficiency of multi-sample RNA-seq assembly.

Main Methods:

• Incorporation of consensus calling at low-level features within the Ryūtō framework.
• Utilizing multi-sample RNA-seq data for improved transcriptome reconstruction.
• Leveraging incomplete references to enhance assembly precision.

Main Results:

• Stable improvements in assembly accuracy observed with as few as three replicates.
• Ryūtō outperforms competing methods, offering a tunable sensitivity-precision balance.
• Demonstrated benefits for differential gene expression analysis.
• Consistent assembly improvement across various conditions and time series, independent of filter settings.

Conclusions:

• The extended Ryūtō tool effectively reconstructs consensus transcriptomes from multiple RNA-seq datasets.
• Ryūtō provides a superior and adjustable sensitivity-precision trade-off compared to existing approaches.
• The tool's ability to use references significantly boosts precision, aiding downstream analyses like differential expression.