search

rrwick / Porechop

adapter trimmer for Oxford Nanopore reads
GNU General Public License v3.0
323 stars 124 forks source link

Required two barcodes set as default? #34

Closed timacwalker closed 6 years ago

timacwalker commented 6 years ago

Hi there,

After running porechop with default settings and --verbosity 3 I noticed that it seems a lot of my reads aren't being classified because there isn't a threshold match at both ends. As I understand it, in the default settings this shouldn't be happening. Have I understood the info below correctly? If I have, is there a way to stop this?

Thanks,

Tim.

...ACGGCTCATCGGGACTAAAAGATCACCTGCATCCACGAGTGACATGACGTCGTCTATACGGGCTCAGCAACGCCCATCCGTTTGGAGCCTCATCAGGTAATTCTTTACTCCCTCAGGATGCAAGCCTATATATAAGGTTTGCGGTGCGGG Barcodes: start barcodes: BC02 (78.6%), BC01 (48.0%) end barcodes: BC02 (64.0%), BC01 (59.3%) best start barcode: BC02 (78.6%) best end barcode: BC02 (64.0%) barcode call: none

rrwick commented 6 years ago

Hi Tim,

You've interpreted this correctly, so it's not immediately obvious to me what's going on.

One thought jumps to mind: are you running Porechop on an Albacore output directory? In the last version I added logic such that if Porechop and Albacore disagree on the barcode, then Porechop won't bin the read. So perhaps your example read didn't get an Albacore bin (wouldn't surprise me if the best hit was only 79%) and so Porechop didn't bin it either.

To make this more clear, I've just added the Albacore core bin to the verbosity level 3 output in Porechop's development branch. You're welcome to try that if you want.

If you're not running Porechop on an Albacore directory, then I don't know what the problem is. Let me know and I can dig deeper.

Ryan

rrwick commented 6 years ago

This is an old issue, so I'm going to close it now. But please let me know if you're still experiencing unresolved issues and we can reopen it!

Ryan