Identifying flanking LTR pairs

[bryan-n-arch]

Hello,

I've used EDTA to successfully annotate and mask my plant genome (via RepeatMasker). However, I am also interested in the actual flanking LTR pairs for each LTR retrotransposon.

I know that LTR_finder and LTR harvest report these on their own. By running them individually on a segment of my genome, I'm able to only regenerate some of these pairs (maybe less than 10% of the total unique types found by EDTA). And furthermore, many of them do not match the reported positions found by running the full EDTA pipeline.

What would be the best way to find the corresponding LTR pairs for each LTR subfamily reported?

Much appreciated,

Bryan

[oushujun]

Dear Bryan,

Currently the full EDTA annotation pipeline is aggressively removing nested insertions and reducing the redundancy, which will result in a fragmented TE annotation. This could be the reason that you find the annotated TEs are not giving you the precise LTR region. I am working on resolving this issue, but alternatively, you can use the intact TE annotation file which contains the information you are looking for. It's located in the *EDTA.raw/ folder with the *intact.fa.gff3 naming.

Please let me know if this answers your question. Thank you for testing EDTA.

Best,
Shujun

[bryan-n-arch]

Hi Shujun,

I don't have this file in my "*EDTA.raw" directory (or at least I'm not seeing it) -- perhaps because I used an older version? Our library was built a few months ago.

I do, however, have a "*.pass.list.gff3". It seems like it contains the left and right flanking pairs reported by LTR retriever. Can I rely on this to derive the flaking LTR pairs?

I appreciate the help.

Cheers,

Bryan

[oushujun]

Hi Bryan, Yes this is the file you are looking for. I recommend you to update EDTA because of a recent major update, which will improve thd overall annotation accuracy. However, for intact LTR it is likely the same. Best, Shujun

…

On Tue, Dec 10, 2019, 2:45 PM Bryan Naidenov ***@***.***> wrote: Hi Shujun, I don't have this file in my "*EDTA.raw" directory (or at least I'm not seeing it) -- perhaps because I used an older version? Our library was built a few months ago. I do, however, have a "*.pass.list.gff3". It seems like it contains the left and right flanking pairs reported by LTR retriever. Can I rely on this to derive the flaking LTR pairs? I appreciate the help. Cheers, Bryan — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#32?email_source=notifications&email_token=ABNX4NG3QH2XLLYACFZ7NB3QX75WZA5CNFSM4JXS2GTKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEGQ4MSA#issuecomment-564250184>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABNX4NB6DBLZEWMJ5JER7MLQX75WZANCNFSM4JXS2GTA> .