Fig. 9

Results for the two Illumina short-read datasets. Precision, recall and F-scores for the species-level analysis based on a minimum threshold of 0.001% of the total reads for A Illumina ATCC MSA-1003 and B Illumina Zymo D6300. Species-level relative abundance estimates for C Illumina ATCC MSA-1003 and D Illumina Zymo D6300. The theoretical distributions are shown on the left and are based on the manufacturer’s specifications. The read counts for all species-level false positives were grouped in a category labeled ‘Other’. Asterisks signify methods that failed the chi-squared goodness of fit test