Overview of the REAPR pipeline. (a) The input is a BAM file of read pairs mapped to the assembly. (b) Statistics are calculated at each base of the genome: (i) Read coverage per strand, and any perfect and uniquely mapped read coverage is incorporated; (ii) The type of read coverage on the forward (upper plots) and reverse (lower plots) strand: proportion of reads that are properly paired (red), orphaned (green), and in the wrong orientation or exceed the fragment size range (not shown); (iii) The number of reads soft-clipped at each base; (iv) The fragment coverage, determined by the properly paired reads; (v) FCD error, taking into account the presence of a gap. Boxed are: FCD calculation at a given base. The fragments covering that base, shown in red, are used to construct a fragment depth plot (red). The FCD error is the area (grey) between the observed (red) plot and ideal plot (green). Since no read can map to a gap in the assembly, the calculation is corrected when a gap is present. (c) The statistics at each base are used independently to assign a score to each base of the assembly and also to break the assembly at scaffolding errors.