BUSCO gene hits

[TCHeaven]

I have been trying to provide hits to blobtools via blastx search of BUSCO gene regions vs Uniprot, however all contigs are being assigned to no-hit despite there being taxid hit info in the input file:

`ptg016977l:1-7766 121845 223.4 ptg016977l:1-7766 tr|A0A3Q0J621|A0A3Q0J621_DIACI 93.3 119 8 0 5149 4793 25 143 4.1e-54 223.4

ptg016977l:1-7766 121845 161.4 ptg016977l:1-7766 tr|A0A3Q0J621|A0A3Q0J621_DIACI 85.4 82 12 0 1757 1512 179 260 1.9e-35 161.4

ptg004722l:33979-35044 28743 468.8 ptg004722l:33979-35044 tr|A0A3Q2GQF1|A0A3Q2GQF1_CYPVA 65.5 354 106 1 1 1062 8 345 7.7e-129 468.8

ptg008147l:570-8350 7740 451.1 ptg008147l:570-8350 tr|A0A8K0EU27|A0A8K0EU27_BRALA 42.7 536 294 5 3707 5278 579 1113 1.2e-122 451.1
`

A BUSCOgenes taxrule was mentionaed in the recent workshop but I can't find reference to this anywhere. I am wondering if I need to remove the text follow in ':' characters in columns 1 and 4, however the test data provided in the recent workshop had text in addition to the contig names following ':' and provided hits to the plot without issue:

`ptg000043l:272955-274207=1275837at2759=single 1903189 553 ptg000043l:272955-274207=1275837at2759=single tr|A0A8H3G8J7|A0A8H3G8J7_9LECA 93.4 316 15 3 1 315 1 311 2.01e-196 553

ptg000043l:272955-274207=1275837at2759=single 560253 553 ptg000043l:272955-274207=1275837at2759=single tr|A0A8H6CAM3|A0A8H6CAM3_9LECA 93.4 316 15 3 1 315 1 311 2.85e-196 553

ptg000043l:272955-274207=1275837at2759=single 112416 553 ptg000043l:272955-274207=1275837at2759=single tr|A0A8H6G095|A0A8H6G095_9LECA 93.4 316 15 3 1 315 1 311 2.85e-196 553

ptg000043l:272955-274207=1275837at2759=single 172621 549 ptg000043l:272955-274207=1275837at2759=single tr|A0A8H3FIP3|A0A8H3FIP3_9LECA 93.0 316 16 3 1 315 1 311 1.45e-194 549`

[rjchallis]

When blobtools sees the format in the workshop example, it should automatically recognise it as the output of the diamond blastp step from the blobtools pipeline and treat it as a blastp file and parse the details in the sequence_id columns accordingly, in this case the tax rule provided really only changes the name of the output fields so buscogenes acts as a label and setting --taxrule buscogenes is the same as explicitly setting the blastp tax rule and giving it an alternate name with --taxrule blastp=buscogenes.

The sequence IDs in your file look to be missing the =1275837at2759=single part that adds the busco gene information so the import is treating the sequence ID as being ptg016977l:1-7766, as you thought, removing everything after the : should fix the problem as then the sequence IDs will match the sequence IDs from the assembly fasta file