Skip to main content
Fig. 3 | Genome Biology

Fig. 3

From: VirStrain: a strain identification tool for RNA viruses

Fig. 3

A The accuracy comparison of 12 tools. There are 100 sets of simulated reads for single-strain datasets and 100 for multi-strain datasets. For each set of multi-strain simulated reads, there are two strains with 100X and 10X coverage, respectively. B The bubble plot of the predicted abundance distributions for 100 simulated SARS-CoV-2 two-strain datasets. The center of each circle represents the relative abundance of the two strains output by one tool. When a tool produces the same abundance distribution on multiple datasets, we represent the identical output using a circle, whose size represents the number of those datasets. “Truth” refers to true relative abundance of the 2 strains in each dataset, which is calculated by normalizing the sequencing depth (100X and 10X). Its circle contains 100 datasets (samples). Many circles have centers with the x-coordinate being 0, meaning that these tools only output one strain

Back to article page