# Introducing gold-standard essential gene datasets for Pseudomonas aeruginosa to enhance Tn-Seq analyses

## TL;DR

This paper introduces gold-standard essential gene datasets for Pseudomonas aeruginosa to improve the reliability and consistency of Tn-Seq analysis methods.

## Contribution

The study provides the first gold-standard essential gene datasets for Pseudomonas aeruginosa to evaluate and compare Tn-Seq analysis tools.

## Key findings

- Gold-standard datasets of 84 and 115 essential genes for Pseudomonas aeruginosa were established.
- Analysis methods like TRANSIT2 HMM and FiTnEss showed retrieval rates of gold-standard genes ranging from 0% to 100%.
- The datasets enable comparative evaluation of Tn-Seq tools, promoting standardization and reproducibility.

## Abstract

Transposon Sequencing (Tn-Seq) is a high-throughput technique that utilizes transposon mutant libraries to assess gene fitness or essentiality under specific conditions potentially identifying novel therapeutic targets. However, the diversity of statistical methods, bioinformatics tools, and parameters complicates the selection of the most appropriate and reliable analysis pipeline for a given dataset. A significant limitation of existing studies is the absence of a gold-standard set of essential genes (EGs) for evaluating the analysis process. Relying on the original study as a gold-standard is suboptimal, as these results may have been obtained using non-optimal tools. Here, we introduce reliable EG datasets for Pseudomonas aeruginosa to enhance Tn-Seq analyses. By utilizing literature data and sequencing of six samples from PA14 Wild-Type (WT) and PA14 OprD-deficient (ΔoprD), grown in LB medium, we compared EG lists generated by several statistical methods of TRANSIT2 and by the FiTnEss tools. We established a reference dataset of 84 genes found in P. aeruginosa and another gold-standard set composed of 115 genes specific to PA14 grown in LB. Our findings revealed that depending on the analysis method used, retrieval rates of gold-standard genes ranged from 0% to 100%. The Hidden-Markov Model (HMM) method available in TRANSIT2 identified approximately 90% of gold-standard EGs, while FiTnEss identified up to 100%. This study addressed a critical gap in the field by providing gold-standard sets of EGs, enabling comparative evaluation of Tn-Seq analysis methods to help researcher select the most suitable bioinformatics pipeline for a given Tn-Seq dataset. We anticipate that our results will facilitate Tn-Seq analysis comparisons, harmonize P. aeruginosa-related studies, promote standardization and enhance reproducibility. Ultimately, this will lead to more reliable identification of EGs and potential therapeutic targets in P. aeruginosa, advancing our understanding of this important pathogen.

Tn-Seq analyses are a crucial resource for understanding gene function and identifying potential new therapeutic targets. However, during our analyses, we encountered vast array of available tools and disparity in results reported in the literature. Therefore, we conducted a comparative evaluation of bioinformatics tools based on two gold-standard datasets of essential genes for Pseudomonas aeruginosa that we established. This approach enables researchers performing Tn-Seq analysis to assess the quality of their results, thereby promoting consistency and harmonization across studies.