TSSAR

TSSAR annotates transcription start sites (TSS) de novo from differential RNA sequencing (dRNA-seq) data by modeling read-start count distributions to detect significantly enriched primary transcripts.

Key Features:

• Automated Annotation: Performs automated de novo identification of TSS from dRNA-seq data using statistical models tailored to dRNA-seq characteristics.
• Statistical Foundation: Models read counts starting at genomic positions within transcriptionally active regions with Poisson distributions whose parameters depend on local expression strength and models differences between two dRNA-seq libraries using the Skellam distribution.
• Performance and Validation: Was evaluated on publicly available dRNA-seq datasets and reproduced 74 experimentally validated TSS in Helicobacter pylori confirmed by RACE or primer extension.

Scientific Applications:

• De novo TSS annotation: Enables unbiased genome-wide annotation of transcription start sites from dRNA-seq data.
• Transcriptome dynamics monitoring: Facilitates analysis of transcriptome plasticity and dynamics under different stimuli and growth conditions.
• Gene regulation studies: Supports investigations into mechanisms of gene regulation by providing high-confidence TSS locations.

Methodology:

TSSAR models read-start counts at genomic positions with Poisson distributions whose parameters reflect local expression strength and models differences between two dRNA-seq libraries with the Skellam distribution to identify significantly enriched primary transcripts.