Improve the performance in Local blast
I set up a local blast and I have a large sets of queries and a large database. The search takes minimum 15 minutes using blastn- short.
I am using python as a parser and I would like to ask if there is any way to improve the performance. Parallel computing, multithreading or any thoughts? I know about the parameter num_threads, but is it the optimal solution?
Thank you
• 3,244 views
•
link
updated
by
Antonio R. Franco
•
written
by
Mona_AAA •
•
written
by
Mona_AAA •
1 answer
This biostar thread is what you are looking for
• 1 views
•
link
written
by
Antonio R. Franco
Log in to answer this question.
More posts like this
-
Why is local BLAST so slow? (yet another one of these questions)
written by mykle hoban •
-
BioPython BLAST unusable?
written by John •
-
Optimum setting for local blastp for ~10K sequences
written by sodiumnitrate •
-
Command-line input for 'short queries' in blastp using Geneious
written by jan_raschke •
-
searching short peptide sequences against a large database
written by h.mendel •
-
BLAST+ vs Web BLAST "Automatically adjust parameters for short input sequences" command line arg
written by oostin623 •
-
parameter in blastn to limit the output to top 20 best hits
written by J.F.Jiang
-
Blastn Segmentation Fault
written by Srihari •
-
Blastn Runtime 606 Queries
written by W Langdon •
-
Query Vs. Target Using Blast2
written by Guidobot •
Are you using a single computer or a cluster? Can you provide details about the hardware? If you have a standalone computer then num_threads may be the only practical option.
Yes, I'm working on my computer.
What sort of hardware do you have (1 or > 1 CPU, RAM)? With a single CPU your best bet is to use multiple threads for blast.
Unless his CPU supports at least 8 threads, he has enough RAM and fast IO, in which case it would probably be better to run two num_threads 4 jobs in parallel instead of just one num_threads 8 job..