StringTie

StringTie assembles and quantifies full-length transcripts from RNA-Seq alignments to reconstruct transcriptomes and estimate gene expression levels.

Key Features:

• Network Flow Algorithm: Uses a novel network flow algorithm (from optimization theory) to optimize transcript assembly from RNA-Seq alignments.
• Optional De Novo Assembly: Includes an optional de novo assembly step to recover transcripts beyond those represented in the input alignments.
• Comprehensive Transcriptome Reconstruction: Demonstrated superior reconstruction compared with Cufflinks, IsoLasso, Scripture, and Traph; for example, assembled 10,990 transcripts from a human blood dataset of 90 million reads, a 53% increase over Cufflinks.
• Accuracy in Expression Level Estimation: Provides more accurate gene and transcript expression level estimates relative to other assemblers.
• Performance Efficiency: Runs faster than other assembly software on tested datasets.

Scientific Applications:

• Genomics: Applied to detailed transcriptome analysis in genomics for reconstruction and annotation of transcripts.
• Molecular Biology: Used to investigate complex gene expression patterns and alternative splicing events in molecular biology studies.
• Personalized Medicine: Employed for transcript-level expression profiling and novel transcript discovery relevant to personalized medicine.

Methodology:

Implements a network flow algorithm on RNA-Seq alignments with an optional de novo assembly step and has been evaluated on simulated and real datasets.