BLAST+ strange results

[nupurgupta]

I am running command line blastn with a local copy of a microbial database. My command looks like

blastn -query query.txt -db 16SMicrobial -out newres.txt -wordsize 11 -evalue 10 -gapextend 2 -gapopen 5 -penalty -3 -reward 1 This gives me 81 hits, they all have the same score such as

Sequences producing significant alignments: Score (Bits) E Value

ref|NR042397.1| Dokdonella fugitiva strain : A3 16S ribosomal R... 28.2 6.3
ref|NR041093.1| Streptomyces rubiginosohelvolus strain NBRC 129... 28.2 6.3
ref|NR041539.1| Kitasatospora kazusensis strain SK60 16S riboso... 28.2 6.3
ref|NR041538.1| Kitasatospora saccharophila strain SK15 16S rib... 28.2 6.3
ref|NR_044150.1| Streptomyces omiyaensis strain NRRL B-1587 16S ... 28.2 6.3
...

Isn't that strange?

Further, I repeated the query(same parameters) on the NCBI website, and that gave 162 hits. e.g.

NC017551.1 Leptospira interrogans serovar Lai str. IPAV chromosome chromosome 1, complete sequence 398 398 100% 1e-108 100%
NC008463.1 Pseudomonas aeruginosa UCBPP-PA14 chromosome, complete genome 398 398 100% 1e-108 100%
NC_002516.2 Pseudomonas aeruginosa PAO1 chromosome, complete genome 398 398 100% 1e-108 100%

I tried to compare the two, e.g. I looked for "Kitasatospora " from the former in the latter set of results, and could not find any matches. What am I doing wrong ?!

[chris]

I am running command line blastn with a local copy of a microbial database. My command looks like

blastn -query query.txt -db 16SMicrobial -out newres.txt -wordsize 11 -evalue 10 -gapextend 2 -gapopen 5 -penalty -3 -reward 1 This gives me 81 hits, they all have the same score such as

Sequences producing significant alignments: Score (Bits) E Value

ref|NR042397.1| Dokdonella fugitiva strain : A3 16S ribosomal R... 28.2 6.3
ref|NR041093.1| Streptomyces rubiginosohelvolus strain NBRC 129... 28.2 6.3
ref|NR041539.1| Kitasatospora kazusensis strain SK60 16S riboso... 28.2 6.3
ref|NR041538.1| Kitasatospora saccharophila strain SK15 16S rib... 28.2 6.3
ref|NR_044150.1| Streptomyces omiyaensis strain NRRL B-1587 16S ... 28.2 6.3
...

Isn't that strange?

Not to me. Those scores are very poor and probably the worst that can be reported by BLAST for your database. So, although they are *hits* I doubt they are meaningful.

Further, I repeated the query(same parameters) on the NCBI website, and that gave 162 hits. e.g.

NC017551.1 Leptospira interrogans serovar Lai str. IPAV chromosome chromosome 1, complete sequence 398 398 100% 1e-108 100%
NC008463.1 Pseudomonas aeruginosa UCBPP-PA14 chromosome, complete genome 398 398 100% 1e-108 100%
NC_002516.2 Pseudomonas aeruginosa PAO1 chromosome, complete genome 398 398 100% 1e-108 100%

I tried to compare the two, e.g. I looked for "Kitasatospora " from the former in the latter set of results, and could not find any matches. What am I doing wrong ?!

I suspect your searching against different databases as the NCBI hits don't look like 16S matches to your queries.

[nupurgupta]

Thanks for the perspective. But I downloaded the microbial database from ncbi. ftp://ftp.ncbi.nlm.nih.gov/blast/db/16SMicrobial.tar.gz
. I assumed that was the entire microbial database that online NCBI was also using for BLAST. Maybe not, let me see if I can find a FASTA file for all bacteria/microbes that I can use to build a database.
Thanks very very much. This is a huge help

[chris]

It would help if you stated the name of the database you used on NCBI. Then a judgement can be made as to whether you're comparing like with like.

[nupurgupta]

I was using NCBI blast on microbial data.

http://www.ncbi.nlm.nih.gov/sutils/genom_table.cgi

[drdna]

I am not surprised at all. My guess is that your query has a chance (and very short) match to a conserved sequence within the ribosomal 16s sequence. Have you tried looking at the alignments? I predict that every alignment will be the same.

[nupurgupta]

Yes, you are right! Sorry I am a newbie to this. Have found a bacterial database

ftp://ftp.ncbi.nih.gov/gene/DATA/GEN...haea_Bacteria/

Am going to get it now.