TB-annotator: a scalable web application that allows in-depth analysis of very large sets of publicly available Mycobacterium tuberculosis complex genomes

Tuberculosis continues to be one of the most threatening bacterial diseases in the world. However, we currently have more than 160,000 Short Read Archives (SRAs) of Mycobacterium tuberculosis complex. Such a large amount of data should help to the understanding and the fight against this bacterium. To accomplish this, it would be necessary to thoroughly and comprehensively examine this significant mass of data. This is what TB-Annotator proposes to do, combining a database containing all the diversity of these 160,000 SRAs (at least, SRAs with a reasonable read size and quality), and a fully featured analysis platform to explore and query such a large amount of data. The objective of this article is to present this platform centered on the key notion of exclusivity, to show its numerous capacities (detection of single nucleotide variants, insertion sequences, deletion regions, spoligotyping, etc.) and its general functioning. We will compare TB-Annotator to existing tools for the study of tuberculosis, and show that its objectives are original and have no equivalent at present. The database on which it is based will be presented, with the numerous advanced search queries and screening capacities it offers, and the interest and originality of its phylogenetic tree navigation interface will be detailed. We will end this article with examples of the achievements made possible by the TB-Annotator, followed by avenues for future improvement.