Benchmarking RACA and different Ragout 2 strategies using the simulated human Chr 14 assembly. A target and three reference genomes were simulated by randomly distributing the corresponding number of rearrangements along the branches of a phylogenetic tree. In addition to the normal Ragout 2 run, we performed the “permissive” run in which all target-specific adjacencies are accepted and the “conservative” run in which all target-specific adjacencies are broken. (Left) Misassembly rate between the ordered contigs, computed relative to the complete simulated sequence. (Right) The number of unplaced contigs (missing in the assembled chromosomes). Each data set contains approximately 5000 contigs with a mean length of 18,000 bp.
