1
$\begingroup$

I have been running a couple of blast searches and I am very suprised and somewhat frustrated by the huge variability in job completion time. The specifics of my blast search are as follows:

  • Running blast using webbrowser (chrome) via ncbi
  • 19 nucleotide query
  • database: WGS of one particular sequencing project/bacterial strain (so database is VERY small, in total approx 4.5 MB divided over 60 contigs)
  • parameters are mostly untouched and are automatically optimized for short queries by blast

In one instance, I started the job more than an hour ago and it is still not done. Knowing from experience that this previously hasn't taken that much time, I started a new job in a different browser tab with exactly the same sequence, database and parameters, and now it finished within 20 seconds. What on earth is happening here?

Improve this question
$\endgroup$
3
  • $\begingroup$ Welcome to Bioinformatics.SE! I assume you're running BLAST on the command line, and not using a web browser to connect to NCBI's BLAST website. Are you running this on the computer on your desk, or are you connecting to a remote computer? It's possible the machine you're connecting to is experiencing intermittent disk issues. $\endgroup$
    – Daniel Standage
    Commented Aug 2, 2019 at 15:16
  • $\begingroup$ You mention "tabs" so what service are you using? Please edit your question and tell us exactly what you are running and how. What blast? If you're using a website, which one? If not, what blast executable? On what kind of computer? $\endgroup$
    – terdon ♦
    Commented Aug 2, 2019 at 16:17
  • $\begingroup$ Thanks for the tips, I edited my question a bit providing more information - running blast via NCBI (in both cases) $\endgroup$
    – Sincinkybaju
    Commented Aug 3, 2019 at 9:23

1 Answer 1

Reset to default
1
$\begingroup$

First of all, if you're using an online service like NCBI's BLAST, then the time to completion will depend on the load the service is experiencing. Your job is ut in a queue and will be started when it's its turn. If the service is under heavy load, then its turn might come later than you expect. Then, if the service is under heavy load, even after launching your job, it might take longer to finish because it won't have enough resources to run quickly.

Since you say that the same job finished quickly on a new tab, it sounds like either the first job was last in the queue or there was just a bug that caused it not be launched or something like that. We can't really help with things that resolve themselves like that.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.