Improved transcriptome assembly using a hybrid of long and short reads with StringTie

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Short-read RNA sequencing offers high accuracy but struggles with full-length transcript assembly.
• Long-read RNA sequencing captures full-length transcripts but has a higher error rate, leading to splice site misidentification.

Purpose of the Study:

• To present a new release of StringTie enabling hybrid-read transcriptome assembly.
• To demonstrate the superior accuracy and precision of hybrid assembly over single-read methods.

Main Methods:

• Developed and applied a hybrid-read assembly approach using StringTie.
• Integrated both short-read and long-read RNA sequencing data.
• Validated performance on simulated and real datasets from Arabidopsis thaliana, Mus musculus, and human.

Main Results:

• Hybrid-read assembly with StringTie significantly enhances transcriptome assembly accuracy and precision.
• Achieved more than double the number of correctly assembled transcripts on some datasets compared to single-read methods.
• Demonstrated superior performance over long-read correction prior to assembly, with faster processing times.

Conclusions:

• Hybrid-read assembly using StringTie is a more accurate and efficient method for transcriptome assembly.
• This approach leverages the complementary strengths of short and long reads for improved genomic analysis.
• StringTie provides an open-source solution for advanced transcriptome assembly.