Avaliação de metodologia para análises in silico de dados de sequenciamento de genoma total de Klebsiella pneumoniae resistente a carbapenêmicos

Abstract

The emergence of Klebsiella pneumoniae strains resistant to different antimicrobial classes has increased the concern regarding control and monitoring measures for this microorganism. Whole Genome Sequencing (WGS) has been used in epidemiological studies, public health investigations, as well as for tracking and monitoring outbreaks of multidrug-resistant pathogens. In addition, it allows the evaluation of multiple genes through online tools. The analysis and interpretation of data at a genomic scale is challenging due to the data volume, its complexity and the number of programs involved. The pipeline validation for WGS in clinical microbiology is still poorly described. Thus, the present study aimed to validate a bioinformatics pipeline for the antimicrobial resistance genes identification from carbapenemresistant Klebsiella pneumoniae WGS. Sequences were obtained from a publicly available database, trimmed, de novo assembled, mapped to the K. pneumoniae reference genome and annotated. Contigs were submitted to different databases for bacterial (Kraken 2 and SpeciesFinder) and antimicrobial resistance genes (ResFinder and ABRicate) identifications in order to produce standardized methodologies that allow interlaboratory comparisons. The pipeline performance evaluation was carried out according to traditional metrics of repeatability, reproducibility, accuracy, precision, sensitivity and specificity, adapted to the objective of the study. As a result, 100% repeatability and reproducibility were obtained for the evaluated databases. For the other performance metrics, 99.99% accuracy, 83.53% precision, 75.45% sensitivity and 100% specificity were obtained for the ResFinder database. For the ABRicate database, 99.99% accuracy, 83.78% precision, 75.18% sensitivity and 100% specificity were obtained. Discrepancies in bacterial identification and detection of resistance genes were observed probably due to the difference in databases. Due to the range of bioinformatics tools and databases, validation is of paramount importance. The validation strategy used in our study can be applied in different bioinformatics pipelines and databases to ensure intra and interlaboratory standardization.