Further analogies can be found in protein sequence similarity searching wherein BLAST hits are mapped to domains Arylquin 1 annotated within proteins. The algorithm consists of three phases. i) In the preprocessing phase in which a database of an arbitrary number of species is combined into partitions that are then indexed with the Bowtie2-build program of the Bowtie2 package. Alternatively, pre-built indices can be downloaded from the project site. ii) Alignment is then carried out with Bowtie2 and the taxa are identified with a lowest common ancestor search algorithm. The standard output of this phase is a summary of the found taxa and alignments in the SAMtools format. iii) Unlike other metagenome analysis programs Taxoner can optionally provide a list of genes identified at the species level, along with their predicted functions. It also contains a utility that can produce a summary of the found functions based on the COG-eggNog scheme of functional descriptors and by using a B-tree index. In addition, the read alignments provided in the SAM format can be further processed with other taxon assignment programs such as MEGAN. The development of sequencing technologies has given rise to a number of sequencing platforms in recent years. The performance of read aligners often varies on the basis of reads produced by the various sequencing platforms. We RN-9893 compared the performance of aligner programs on read datasets selected from Staphylococcus aureus sequencing data. BLAST aligners had the highest alignment rate, which is not surprising since BLAST is a sensitive local aligner. MetaPhlAn had the lowest alignment rates, which is, again, expected since MetaPhlAn only aligns reads to a unique subset of the bacterial database. In general, all aligners had the best performance on the 454 dataset, since the 454 reads are longer and thus easier to analyze. All programs performed well at genus level. An important aspect of metagenome analysis is the identification of taxa at the lowest possible taxonomic levels such as species or strains. At the species level Taxoner showed the best performance with the exception of 454 reads where BLAST performed better.