Accurate assembly of multi-end RNA-seq data with Scallop2

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Modern RNA sequencing (RNA-seq) protocols generate multi-end data, where multiple reads from a single transcript share a barcode.
• This long-range information is valuable for phasing complex spliced isoforms, but existing assembly algorithms do not fully leverage it.

Purpose of the Study:

• To introduce Scallop2, a novel reference-based RNA sequencing assembler specifically designed for multi-end data.
• To enhance the accuracy of transcript isoform phasing and assembly using multi-end RNA-seq information.

Main Methods:

• Scallop2 employs a three-step algorithmic core: bridging multi-end reads into single-end phasing paths on a splice graph.
• Refining splice graphs using multi-end reads that do not bridge and integrating phase-preserving decompositions.
• Utilizing multi-end reads to improve splice graph accuracy and isoform phasing.

Main Results:

• Scallop2 demonstrated substantial improvements in assembly accuracy compared to StringTie2 and Scallop.
• Performance was validated on diverse datasets, including Smart-seq3 and Illumina paired-end RNA-seq samples.
• The assembler effectively leverages multi-end read information for enhanced transcript phasing.

Conclusions:

• Scallop2 represents a significant advancement in RNA sequencing data analysis, particularly for multi-end protocols.
• The tool offers improved accuracy for phasing complex spliced isoforms, addressing a current limitation in transcript assembly.
• Scallop2 is a valuable resource for researchers working with multi-end RNA-seq data, enabling more precise isoform resolution.