Figure 6
From: A new strategy for better genome assembly from very short reads
Validation of our strategy using simulated short read datasets. Simulated short read datasets from 41 genomes are used, and for each dataset we select at least three reference genomes. The similarity values between the reference and the target genomes are between 0.80 and 0.92. We compute the ability of the selection criteria to exclude misassembled A-contigs, the accuracy of reliable A-contigs, the accuracy of hybrid assembly and the proportion of DBG contigs extended by reliable A-contigs.