MetaGenomic analysis of short and long reads

Identifying single organisms in environmental samples is one of the key tasks of metagenomics. During the last few years, third generation sequencing technologies have enabled researchers to sequence much longer molecules, but at the expense of sequencing accuracy. Thus, new algorithms needed to be developed to cope with this new type of data. With this in mind, we developed a tool called MetaG. An intuitive web interface makes the software accessible to a vast range of users, including those without extensive bioinformatic expertise.

Evaluation of MetaG’s performance showed that it makes nearly perfect classifications of viral isolates using simulated short and long reads. MetaG also outperformed current state-of-the-art algorithms on data from targeted sequencing of the 16S and 28S rRNA genes. Since MetaG’s output is also supplemented with information about hosts and antibiotic resistances of pathogens, we expect it to be especially useful to the healthcare sector.

Moreover, the outstanding accuracy of the taxonomic assignments will make MetaG a serious alternative for anyone working with metagenomic sequences. MetaG can be accessed at http://bioinformatics.uni-muenster.de/tools/metag/.