An empirical strategy to detect bacterial transcript structure from directional RNA-seq transcriptome data.

Area of Science:

• Bacteriology
• Genomics
• Bioinformatics

Background:

• RNA-sequencing (RNA-seq) is increasingly preferred for bacterial transcriptomics due to its ability to detect cDNA at single nucleotide resolution, providing gene expression and transcript structure information.
• Existing automated software for analyzing large-scale bacterial RNA-seq data, particularly for transcript structure, is limited.

Purpose of the Study:

• To develop and validate an automated method for analyzing directional RNA-seq data to identify bacterial transcript structures.
• To annotate operon organization and identify regulons in Salmonella Typhimurium.

Main Methods:

• Developed an empirical method combined with statistical tests to automatically detect transcriptional start sites (TSSs) and transcriptional termination sites (TTSs) from 54 directional RNA-seq libraries of S. Typhimurium.
• Utilized identified TSSs and TTSs, along with intergenic distance and co-expression data, to comprehensively annotate operon organization.

Main Results:

• Successfully identified 2,764 TSSs and 1,467 TTSs for 1331 and 844 genes, respectively.
• Discriminated 215 putative sigma 38 regulons and 863 potential sigma 70 regulons.
• Provided a comprehensive operon annotation for S. Typhimurium 14028s.

Conclusions:

• Directional RNA-seq can detect transcriptional borders with ±10-20 nucleotide resolution, though single nucleotide resolution may be limited by technical factors.
• The developed automated methods, statistical models, and operon annotation pipeline are applicable to RNA-seq studies in other bacteria.
• The defined transcriptomic features for S. Typhimurium 14028s serve as a valuable resource for comparative analyses with other Salmonella serotypes.