command line BLASTN 2.3.0+ different hits from the web blastN

fznajar · 09-27-2016, 10:12 AM

Hi,
I got different matches using the command line BLASTN 2.3.0+ (- remote) option from the NCBI website BlastN! Has anybody experienced this before?

GenoMax · 09-27-2016, 10:17 AM

There must be a difference in the command line you are using and the defaults on the web blast. Have you looked into that? Web blast is likely using 2.4.x, the latest. So if you want to really compare then upgrade your local blast to the latest as well.

fznajar · 09-27-2016, 10:30 AM

I did not. I found Blast 2.5.x. I will give this a shot. However, the difference in the matches is stark. Could the updates in the database itself make that much difference for ITS sequences in two months?

GenoMax · 09-27-2016, 12:37 PM

Two months is a long time so perhaps that is one factor for sure.

fznajar · 09-28-2016, 03:41 PM

Thanks GenoMax. Best

maubp · 09-30-2016, 09:28 AM

The NCBI BLAST website does a lot of parameter optimisation for you based things like query sequence length - the gap settings in particular are worth checking compared to the command line defaults.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
BLAST+ Blastn - Discard hits below a certain score (length)	kayleighepps	Bioinformatics	1	12-13-2013 05:36 AM
Web Services Blastn Problems	Andrew73	Bioinformatics	0	03-05-2013 02:17 PM
PubMed: VARIANT: Command Line, Web service and Web interface for fast and accurate fu	Newsbot!	Literature Watch	0	12-14-2012 12:40 AM
454 contings blastn no hits found!!!	dgio	Bioinformatics	3	10-22-2012 09:04 AM
Problems with blastn execution from command line	angeloulivieri	Bioinformatics	1	10-03-2012 03:48 AM

09-27-2016, 10:12 AM	#1
fznajar Member Location: Oklahoma Join Date: Jan 2012 Posts: 35	command line BLASTN 2.3.0+ different hits from the web blastN Hi, I got different matches using the command line BLASTN 2.3.0+ (- remote) option from the NCBI website BlastN! Has anybody experienced this before?

09-27-2016, 10:17 AM	#2
GenoMax Senior Member Location: East Coast USA Join Date: Feb 2008 Posts: 7,080	There must be a difference in the command line you are using and the defaults on the web blast. Have you looked into that? Web blast is likely using 2.4.x, the latest. So if you want to really compare then upgrade your local blast to the latest as well.

09-27-2016, 10:30 AM	#3
fznajar Member Location: Oklahoma Join Date: Jan 2012 Posts: 35	I did not. I found Blast 2.5.x. I will give this a shot. However, the difference in the matches is stark. Could the updates in the database itself make that much difference for ITS sequences in two months?

09-27-2016, 12:37 PM	#4
GenoMax Senior Member Location: East Coast USA Join Date: Feb 2008 Posts: 7,080	Two months is a long time so perhaps that is one factor for sure.

09-28-2016, 03:41 PM	#5
fznajar Member Location: Oklahoma Join Date: Jan 2012 Posts: 35	Thanks GenoMax. Best

09-30-2016, 09:28 AM	#6
maubp Peter (Biopython etc) Location: Dundee, Scotland, UK Join Date: Jul 2009 Posts: 1,543	The NCBI BLAST website does a lot of parameter optimisation for you based things like query sequence length - the gap settings in particular are worth checking compared to the command line defaults.