Completed on 8 Jun 2015 by .
If you would like to claim attribution for this anonymous review, please contact Academic Karma.
Login to endorse this review.
Sourced from https://static-content.springer.com/openpeerreview/art%3A10.1186%2Fs13742-015-0076-3/13742_2015_76_ReviewerReport_V4_R1.pdf The manuscript describes a scaffolding tool and its use in scaffolding bacteria,
yeast and a very large plant genome. I think the authors have done a fair job of
exploring the parameter space of the presented tool, and I find the manuscript to
be clearly written and mostly convincing. I like that the authors have coupled the
publication of their fairly straight-forward, albeit useful, scaffolding tool with a new
and interesting data type.
(1) I would recommend putting Figure S1 in the main text, as it is critical to
arguing that minion reads have error free k-mers long enough to be useful for
(2) I am somewhat skeptical that a k-mer based approach is the right one for
scaffolding long, error prone reads. It seems to me that an alignment based
approach would produce better results, as you can be more sure that the
resulting linking of the scaffolds is the result of a true, contiguous, orthologous
alignment. Have the authors explored this approach? I realize there are issues of
efficiency to consider, however, I think for the volumes of minion reads currently
available that this isn’t a big deal. Some nice tools, like marginAlign (see the Jain
et al. paper), make higher quality alignments possible with nano-pore reads.
(3) It would be good to reference and discuss the Loman, Quick, Simpson paper
on assembling E.Coli de novo from nano pore reads, which is available on Bio
Arxiv. Clearly error correction can radically improve the utility of minion reads.