search

WGLab / DeepRepeat

An accurate repeat detection from Nanopore data using deep learning and image techniques
Other
19 stars 4 forks source link

[Suggestion] Adding optional long-read mapper #3

Open tuannguyen8390 opened 2 years ago

tuannguyen8390 commented 2 years ago

Hi there,

As per recent development of the T2T Consortium, I am aware that there is a new long-read mapper called Winnowmap2, it was designed to be more sensitive toward repetitive regions compared with Minimap2 or NGLMR (illustrate in Figure 2 in the publication)

I think DeepRepeat is currently choose Minimap2 as default setting, but wonder if this is worth considering as an optional feature in upcoming release.

Best,

Tuan

liuqianhn commented 2 years ago

@tuannguyen8390 Thank you very much for your suggestions. It is glad that the new mapper can generate better alignment for repetitive regions. Right now, DeepRepeat takes the alignment generated by minimap2 as input. In the future, we will check/integrate/evaluate the alignment generated by Winnowmap2. Feel free to share any of your opinion or your evaluation. Thanks.

tuannguyen8390 commented 2 years ago

Thanks for your reply, I did have some investigation, in bovine, there are a few regions that were marked to be filled with tandem repeat. I've been manually analyzed a few regions and saw that Winnowmap2 did have clearer mapping compared to Minimap2