Figure 2

Dispersity score. Illustration of dispersity measurement. Read pairs linking contigs c₁ and c₂ of lengths n and m respectively are transformed to data tested with the KS-test. (a) Observations from contig c₁ are translated and reflected on the x-axis while observations from contig c₂ are translated. The two sample KS statistic will indicate high similarity in read distribution. (b) Strange placement of linked reads occur. Several explanations are possible. One possible explanation is that contig c₂ is misassembled (chimeric) and another explanation is that c₂ is a correctly assembled contig with small repeated regions solved on assembly level. The repeat might not be present in other contigs from the assembly and therefore, the alignments to these regions are reported as unique. Contig c₂ is however not close to the to contig c₁ on the genome and linked reads fail to place at the non-repeated regions on c₂. The KS test will indicate low similarity