Fig. 1From: A novel semi-supervised algorithm for the taxonomic assignment of metagenomic readsA subset of non-overlapping reads in a cluster. A cluster consists of 16 reads. A subset of 5 non-overlapping reads from r 1 to r 5 covers most of positions from x to y in the reference sequence. Those reads are also mapped with reference sequences G 1,G 2,G 3,G 4 with high bit-scoresBack to article page