jts / nanopolish

Signal-level algorithms for MinION data
MIT License
559 stars 159 forks source link

Problem with eventalign #1128

Closed MarioRinBarr closed 5 months ago

MarioRinBarr commented 7 months ago

Hi,

I am writing to you because I have a problem with nanopolish. When using eventalign it is not able to align most reads, so in the end all reads fail, for one reason or another: [post-run summary] total reads: 69721, unparseable: 0, qc fail: 190, could not calibrate: 8740, no alignment: 60749, bad fast5: 281 Do you know why this is happening?

Thank you, Mario

jts commented 6 months ago

I suspect you are using R10 reads, which nanopolish does not support

MarioRinBarr commented 6 months ago

Uffff, you are right, I am using R10 reads... Do you know if there is a method to transform R10 reads into R9? Thank you very much!

hasindu2008 commented 6 months ago

@MarioRinBarr You may try f5c which an optimised reimplementation of nanopolish which has an r10 model.