Correction to: SLR: a scaffolding algorithm based on long reads and contig classification

Correction to: BMC Bioinformatics (2019) 20:539

https://doi.org/10.1186/s12859-019-3114-9

Following publication of the original article [1], the author reported that there is an error in the original article;

• 1. The figures’ order in HTML and PDF are incorrect.

  In the original article incorrect Fig. 1 is the correct Fig. 4.

  In the original article incorrect Fig. 2 is the correct Fig. 5.

  In the original article incorrect Fig. 3 is the correct Fig. 6.

  In the original article incorrect Fig. 4 is the correct Fig. 1.

  In the original article incorrect Fig. 5 is the correct Fig. 2.

  In the original article incorrect Fig. 6 is the correct Fig. 3.

Fig. 1

Nine figures plotting NGA50 vs Misassemblies. The results of SLR usually can be found in the top-left corner, which can illustrate the advantage of SLR

Fig. 2

Contig classification combines with SSPACE-LR and LINKS

Fig. 3

NGA50 for datasets produced by repeat-aware evaluation framework

Fig. 4

An example of alignment position revision. For an alignment given by the alignment tool, the region [sr₁₁, er₁₁] (region₃) in the long read lr₁ is aligned with the region [sc₁₁, ec₁₁] (region₁) in the contig c₁. Because sr₁₁ < sc₁₁ and LEN(lr₁) − er₁₁ > LEN(c₁) − ec₁₁, it means the region [0, sr₁₁] (region₄) in lr₁ is not aligned with c₁, and the region [ec₁₁,LEN(c₁) − 1] (region₂) is not aligned with lr₁. However, when lr₁ is truely aligned with c₁ and the alignment is reliable, region₄ should be aligned with the region [sc₁₁ − sr₁₁, sc₁₁] in c₁, and region₂ should be aligned with the region [er₁₁, er₁₁ + LEN(c₁) − ec₁₁]. Because of the high sequencing error rate in long reads, the alignment tool usually does not provide accurate alignment start and end positions. Then, SLR sets sc11′ = sc₁₁ − sr₁₁, sr11′ = 0, ec11′ = LEN(c₁) − 1 and er11′ = er₁₁ + LEN(c₁) − ec₁₁. When the alignment is reliable, the region [sc11′, ec11′] in c₁ is aligned with the region [sr11′, er11′] in lr₁

Fig. 5

There are six contigs (c₁,c₂,c₃,c₄,c₅,andc₆) that can be aligned with the long read lr₁. Because c₁ and c₂ are simultaneously aligned with the left end of lr₁, SLR retains only contig c₁ which has the greatest alignment length, and deletes the alignment information between c₂ and lr₁. Because c₅ and c₆ have been simultaneously aligned with the right end of lr₁, we keep only c₅, and delete the alignment information between c₆ and lr₁. Finally, SLR determines the orders and orientations of c₁, c₃, c₄ and c₅, which form a local scaffold

Fig. 6

(a) There are six long reads: lr₁, lr₂, lr₃, lr₄, lr₅, and lr₆. The contigs c₁ and c₂ are aligned with lr₁. c₃, c₄ and c₅ are aligned with lr₂. c₆, c₄ and c₇ are aligned with lr₃. c₇, c₈ and c₉ are aligned with lr₄. c₁₀, c₁₁ and c₁₂ are aligned with lr₅. c₉, c₁₁, c₁₃ and c₂ are aligned with lr₆. We assume that all these alignments are forward, and all contigs are longer than L_ca. (b) Based on the alignment result described in (a), SLR obtains six local scaffolds: ls₁, ls₂, ls₃, ls₄, ls₅, and ls₆. (c) The scaffold graph G₁ is built using all contigs. We find that G₁ is complicated. (d) Based on the contig classification method described in Section 2.2, the contigs can be divided into two categories. Because c₄ is located in the middle position of ls₂ and ls₃ and has two distinct 3′-end neighbours and two distinct 5′-end neighbour contigs, it is identified as an ambiguous contig. c₁₁ is also an ambiguous contig. The remaining contigs are identified as unique contigs. The scaffold graph G₂ is built based on unique contigs and is thus less complicated than G₁

In this correction article the figures are shown correct.