pogorely / ALICE

Detecting TCR involved in immune responses from single RepSeq datasets
GNU General Public License v3.0
25 stars 13 forks source link

OLGA pipeline 'nchar' error #11

Open marcRDM opened 4 years ago

marcRDM commented 4 years ago

Hi guys,

Thank you for developping this tool, very cool approach! I am strugling a bit to understand exactly how the 'iter' and 'nrec' parameters influence the result you get, hence I would like to run the OLGA pipeline as the pgen does not depend on these.

It works fine on your test data and on some of my data but for some of it I am getting this error: "Error in 2:(nchar(str) - 1) : argument of length 0". Looking into the script, I tried playing a bit with the CDR3 length distribution in my input with no success. My data seems very similar to yours in terms of V/J gene usage and CDR3 length distribution.

I can probably dig deeper but I wanted to know if you encountered this before and could save me some time...

Thanks in advance!

pogorely commented 4 years ago

Hi,

1) Please check if everything (CDR3.amino.acid.sequence) is character class, and not factor. Convert it to character and try again. 2) If it did not work, could you please share a sample of your data which is enough to reproduce this error?

Best, Mikhail Pogorelyy