search

iqbal-lab / Mykrobe-predictor

Antibiotic resistance predictions in minutes on a laptop
Other
50 stars 19 forks source link

Inconsistency between Myrkrobe results and assembly #146

Open sachalau opened 6 years ago

sachalau commented 6 years ago

Hello,

I am using Mykrobe to detect resistance on reads from two different staphylococcus aureus isolates that are very closely related and share the same resistance phenotype. Among the results, two mutations conferring resistance to quinolones are detected :

gyrA S84L 194:0:0.99999 grlA/parC S80F 1:0:0.999999 for one

gyrA S84L 336:0:0.99999 grlA/parC S80F 298:0:0.999999 for the other one

However, I assembled the genome an checked the corresponding proteins, and found that both mutations were absent. I aligned the reads on the assembled genome in order to see if I was not missing a mixed population (but the sequencing was performed directly on the strain) but I could not detect any mixed signal at those position. I aligned with bwa.

Have you already observed such problem? Moreover is it usual to have such a good confidence with one read for alt and zero for ref?

Thanks for your help and for the cool software !

iqbal-lab commented 6 years ago

Sorry for the slow reply @Cashalow . I don't think we have seen this before - @Phelimb ?

sachalau commented 6 years ago

Thanks for your answer ! Do you want me to provide some fastq data associated with this problem so you can check you can replicate my problem ?

iqbal-lab commented 6 years ago

Probably a good idea. I'm afraid I'm about to go on vacation for a week. Will get back to you ASAP after. In the meantime can you give the precise version of mykrobe, what operating system and the output json please

sachalau commented 6 years ago

I'm using the version on bioconda which is mykrobe v0.5.6-0-gbd7923a-dirty, on Linux Can I send you the rest of the data through your professional email ?

iqbal-lab commented 6 years ago

Yes but about yo get on plane for vacation. Back in 10 days. Apologies for delay

iqbal-lab commented 6 years ago

I've realised this is still open @sachalau - many apologies. I think I/we probably need to send the data. My email is zi AT ebi.ac.uk - send me a mail and we'll figure out how to get it to us.

sachalau commented 6 years ago

I sent you the data a while ago from my professional mail I think. Thanks for getting back to me.

iqbal-lab commented 6 years ago

Argh. Any hint as to your email address or email title?