Resumo

The multilocus sequence typing (MLST) method allows the identification of bacterial lineages based on sequencing of housekeeping genes, which can be used for public health and global epidemiology studies when integrated into bioinformatic tools. Many public databases are available on the internet for querying, depositing and comparing MLST data of bacterial species. However, many difficulties exist for a quick comparison due to the independence of these bioinformatic systems, differences in MLST protocols and the absence of fixed patterns among the biological data already stored. The main goal of the present work was the development of MLSTBRNET, a web-based system for storage and analysis of MLST data focusing on Brazilian bacterial samples with integration into external curated MLST database systems such as “mlst.net” and “mlst pasteur” for easy sequence comparison. This web system was developed with Ruby and Javascript programming languages, and uses the “Ruby on Rails” framework and the “Jquery” library for the dynamic page generation. Also, the “BioRuby” library was employed for parsing all biological string data. The data storage is managed by "Sqlite3". The system allows the typical querying methods: by ID, sequence type, nucleotide sequence or by Blastn tool. The search results are compared with the local system database indicating if the queried organism corresponds a new isolate. New species with their respective MLST protocols can be created and managed by allowed users and the new isolate data can be deposited without depending on a webmaster. The data analysis can be performed by querying on the local database and also by the web scrapping on “mlst.net” and “mlst pasteur”. The MLSTBRNET provides a robust structure, with flexibility for data storage, comparison and analysis with independent management performed by registered user-researchers. This innovation integrates with other MLST network databases accelerating sample analysis for epidemiologic studies.