Filter field without any 'PASS' flag

[gbdias]

Hi @adamewing,

• I am trying to use tldr on a public Drosophila melanogaster Nanopore dataset generated in this paper: https://www.g3journal.org/content/8/10/3143.short.

• This dataset is ~30x coverage and has the following length distribution.

  

• I used a high quality reference genome from a different D. mel strain as the mapping reference. https://www.nature.com/articles/s41588-017-0010-y

• My tldr results include over 1000 predictions, but none of them has a 'PASS' flag in the filter field. The flags are listed below.

  gd98309@ss-sub2 tldr_430029$ cut -f23 iso1_ont_sort.table.txt | sort | uniq -c
    1 Filter
  1087 LeftFlankSize,RightFlankSize,UnmapCoverNA,NoTEAlignment,NonRemappable,ShortIns
    8 LeftFlankSize,RightFlankSize,UnmapCoverNA,NoTEAlignment,NonRemappable,ShortIns,ZeroMapQ
   13 LeftFlankSize,UnmapCoverNA,NoTEAlignment,NonRemappable,ShortIns
    3 LeftFlankSize,UnmapCoverNA,NoTEAlignment,NonRemappable,ShortIns,ZeroMapQ

• Do you have any advice on how to treat these predictions that do not pass the filter?

Thanks

[adamewing]

Hi, thanks for your interest. This looks a lot like issue #7 which had to do with the formatting of the TE reference .fasta (-e/--elts). As of the commit referenced in that issue tldr will enforce the ">Family:Subfamily" formatting. Does that help?