Scallop-LR

Scallop-LR assembles transcripts from long-read sequencing data (e.g., PacBio) using a reference-based approach to improve identification of known and novel isoforms.

Key Features:

• Long-Read Optimization: Handles long reads generated by PacBio and other third-generation sequencing platforms, addressing incomplete cDNA synthesis and platform-specific sequencing length limitations that produce partial transcript reads.
• Enhanced Transcript Assembly: Integrates long-read-specific algorithms to improve transcript reconstruction and isoform resolution from error-prone long-read data.
• Quantitative Performance: In comparative analyses of 26 PacBio datasets from the Sequence Read Archive (SRA), identified 2,100–4,000 more known transcripts and 2.4–4.4 times more potentially novel isoforms in human datasets versus Iso-Seq Analysis, and demonstrated superior recovery compared to StringTie.

Scientific Applications:

• Transcriptome reconstruction and isoform discovery: Produces more complete assemblies for identifying known and novel isoforms to study gene expression complexity and regulation.
• Comparative transcriptomics: Applied to datasets including human and mouse to explore transcript diversity across species.

Methodology:

Reference-based transcript assembly from long-read sequencing using long-read-specific algorithms to mitigate high error rates and incomplete cDNA synthesis.