Computationally Efficient Assembly of Pseudomonas aeruginosa Gene Expression Compendia

Area of Science:

• Microbiology
• Bioinformatics
• Genomics

Background:

• Pseudomonas aeruginosa is a significant opportunistic pathogen responsible for severe infections, particularly in immunocompromised individuals and cystic fibrosis patients.
• Effective treatment of P. aeruginosa infections is challenging due to its intrinsic and acquired resistance mechanisms.
• Publicly available RNA sequencing (RNA-seq) data offer a rich resource for understanding P. aeruginosa gene expression but require standardized processing.

Purpose of the Study:

• To aggregate and reanalyze thousands of publicly available P. aeruginosa RNA-seq datasets.
• To develop and validate a robust computational framework for processing and normalizing microbial gene expression data.
• To create a valuable resource for hypothesis generation and testing in P. aeruginosa research.

Main Methods:

• Uniform processing of raw RNA-seq data using the Salmon pseudoaligner.
• Development of filtering criteria to exclude low-quality or aberrant samples.
• Normalization of gene expression data using the ratio-of-medians method.
• Analysis of mapping effects using noncognate reference genomes (PAO1 and PA14).
• Development of an algorithm for incorporating new SRA data.

Main Results:

• Significant improvement in gene expression correlations for co-regulated genes after filtering and normalization.
• Demonstration of the impact of using noncognate reference genomes on expression profiles.
• Creation of a high-quality, normalized compendium of P. aeruginosa transcriptional profiles from 2,333 samples.
• Development of a scalable workflow applicable to other microbial species.

Conclusions:

• The developed framework provides a scalable method for analyzing large-scale microbial RNA-seq data.
• The processed compendia facilitate cross-experiment, cross-strain, and cross-condition analyses of P. aeruginosa.
• This resource aids in understanding P. aeruginosa pathogenesis and developing novel therapeutic strategies.
• The presented workflow can be adapted for the analysis of gene expression data from other microbial pathogens.