search

nanoporetech / bonito

A PyTorch Basecaller for Oxford Nanopore Reads
https://nanoporetech.com/
Other
389 stars 120 forks source link

Accuracy inconsitency is observed #286

Open waka45 opened 2 years ago

waka45 commented 2 years ago

Hi,

We bought the ONT nanopore sequencing machine MinION, and tested using fix-length library (e.g. Eco.li). A confusing result was observed during the sequencing, we found a significant drop of accuracy from about 97% to 80%. in the same channel. Do you have any idea on it? I attached the figure of accuracy vs read (sort by time) below for your information.

image

Thanks very much!

iiSeymour commented 2 years ago

Hi @waka45

Did this occur only on channel 507? Also can you colour by mux?

waka45 commented 2 years ago

Hi @iiSeymour, please see pic below, seems the accuracy difference could be attributed to mux switch..

Why this happend and what setting I can use to make sure that swithch of mux won't decrease my accuracy? Beside, the accuracy drop after mux switch is not occasional. I use kit12, R10.4.1 and MinION.

Thanks and look forward to your reply.

image