the input device is not a TTY

[cabeaudoin]

Hello and Happy Friday,

I installed the docker for psortb, and it works great on the command line; however, when I ran it in a bash script to screen a bunch of proteins (~7,000), it outputs this error

"the input device is not a TTY"

I tried looking around Stackoverflow for docker issues, but it seems it has something to do with the "-it" when installing/running the docker (obviously I'm not really sure what all this docker business is about). Either way, I was just wondering if you might have any way of working around this issue.

I've attached the bash script here for reference. Any help or information would be greatly appreciated. Thank you for your time.

Best,
Chris
run_psortb.txt

[gemmahoad]

Hi Chris,

I have just installed the docker image on Ubuntu and it works ok from a bash script with 1 fasta sequence as input. Have you had any success running the bash script with different/smaller fasta files?
(I'm trying to rule out memory issues).

Thanks,
Gemma

[cabeaudoin]

Hey Gemma,

Thanks for your quick reply. The fasta files that I'm putting in are only one protein sequence long (usually ~200 amino acids). I read something that it might be because I'm piping(...ish) results into the command. Would you be able to try it in a for loop like in the attached file in the first message? Like, within the for loop: store a sequence in a temp file and try using that?

I really appreciate all of your help! Thank you for your time.

Best,
Chris

[gemmahoad]

Hi Chris,

I tried your script and I also saw the same error messages. It's not the psortb program at fault - your bash script loop doesn't work (you can debug by naming the split fasta files as $counter.fasta and seeing what files are created).

Have you tried running psortb with the fasta file of 7000 sequences? i.e.
/home/chris/software/psortb1/psortb_commandline_docker/psortb -i protein_list.fasta -r psortb_out --positive

This will create a concatenated file of psortb results in the psortb_out directory.

Thanks,
Gemma

[hahahaL]

Hey Gemma,

Thanks for your quick reply. The fasta files that I'm putting in are only one protein sequence long (usually ~200 amino acids). I read something that it might be because I'm piping(...ish) results into the command. Would you be able to try it in a for loop like in the attached file in the first message? Like, within the for loop: store a sequence in a temp file and try using that?

I really appreciate all of your help! Thank you for your time.

Best, Chris

You can try change the psortb command,vim psortb and delete the docker's parameters -t will help you