Get chromosome number from BLAST results

[logicthief]

Hi everyone,

I am a beginner in bioinformatics, could anyone tell me how to extract the chromosome number of each BLAST hit for a bunch of query sequences? I looked at the "hit table" but only found the start and end loci of each hit, and I knew there's chr info in standard BLAST report (txt format), it's like below, hard to manipulate on a large scale:

Query=
Length=59


Score E
Sequences producing significant alignments: (Bits) Value

ref|NT_039716.7| Mus musculus strain C57BL/6J chromosome X ge... 93.5 2e-17
ref|NW_001035178.1| Mus musculus strain mixed chromosome X ge... 93.5 2e-17

ALIGNMENTS
>ref|NT_039716.7| Mus musculus strain C57BL/6J chromosome X genomic contig, MGSCv37
C57BL/6J
Length=15097629

Score = 93.5 bits (50), Expect = 2e-17
Identities = 57/60 (95%), Gaps = 1/60 (2%)
Strand=Plus/Plus

Query 1 TACC-CTGTAGGGTTGATAAGCTTATGTTCACTATAACAATTAACACATTTGCCATTGAC 59
|||| ||||| | |||||||||||||||||||||||||||||||||||||||||||||||
Sbjct 1486960 TACCTCTGTAAGATTGATAAGCTTATGTTCACTATAACAATTAACACATTTGCCATTGAC 1487019


>ref|NW_001035178.1| Mus musculus strain mixed chromosome X genomic scaffold, alternate
assembly Mm_Celera 232000009784844, whole genome shotgun
sequence
Length=8237495

Features flanking this part of subject sequence:
323432 bp at 5' side: uncharacterized protein LOC211208
337262 bp at 3' side: uncharacterized protein LOC73934

Score = 93.5 bits (50), Expect = 2e-17
Identities = 57/60 (95%), Gaps = 1/60 (2%)
Strand=Plus/Plus

Query 1 TACC-CTGTAGGGTTGATAAGCTTATGTTCACTATAACAATTAACACATTTGCCATTGAC 59
|||| ||||| | |||||||||||||||||||||||||||||||||||||||||||||||
Sbjct 577313 TACCTCTGTAAGATTGATAAGCTTATGTTCACTATAACAATTAACACATTTGCCATTGAC 577372
I am wondering whether the ASN file would contain such information (it seems not human readable?). If not, the only way I can think of is to extract chr no. from the standard report by perl or grep (another problem is I don't know how to write perl scripts)... Thanks a lot!

[tomc]

blast knows nothing of chromosomes so it they are in a blast report it is because you happened to have a blast database constructed from fasta file with Chr info in the their defline ( the line above the sequence that begins with ">" )
I think your choices are going to be either,
extract it from the report
(which can be messy as there are also many standards for writing deflines as well)

or find/make blast databases that are already per chromosome then any hits are
to the chromosome you are blasting.

[logicthief]

Tomc, thank you. Making such a database seems too challenging for me....

[logicthief]

Update.
just got the answer from a nice NCBI staff:
choose NCBI genome(chromosome) as the database for web megablast, specify the organism (here I used mouse genome), then you get a lot NC_'s (and NT's NW's etc.) in the hit table, the NC is complete chromosome, NC_000067 stands for chr1, NC_000085 for chr19, NC_000086/87 for X/Y (I've left off the current .version number for these accessions). Besides, you can get the coordinates of alignment on corresponding chromosome by looking at column 9 and 10. For hit table format, click here(http://www.ornl.gov/sci/techresource...me/blast.shtml).

Hi everyone,

I am a beginner in bioinformatics, could anyone tell me how to extract the chromosome number of each BLAST hit for a bunch of query sequences? I looked at the "hit table" but only found the start and end loci of each hit, and I knew there's chr info in standard BLAST report (txt format), it's like below, hard to manipulate on a large scale:

Query=
Length=59


Score E
Sequences producing significant alignments: (Bits) Value

ref|NT_039716.7| Mus musculus strain C57BL/6J chromosome X ge... 93.5 2e-17
ref|NW_001035178.1| Mus musculus strain mixed chromosome X ge... 93.5 2e-17

ALIGNMENTS
>ref|NT_039716.7| Mus musculus strain C57BL/6J chromosome X genomic contig, MGSCv37
C57BL/6J
Length=15097629

Score = 93.5 bits (50), Expect = 2e-17
Identities = 57/60 (95%), Gaps = 1/60 (2%)
Strand=Plus/Plus

Query 1 TACC-CTGTAGGGTTGATAAGCTTATGTTCACTATAACAATTAACACATTTGCCATTGAC 59
|||| ||||| | |||||||||||||||||||||||||||||||||||||||||||||||
Sbjct 1486960 TACCTCTGTAAGATTGATAAGCTTATGTTCACTATAACAATTAACACATTTGCCATTGAC 1487019


>ref|NW_001035178.1| Mus musculus strain mixed chromosome X genomic scaffold, alternate
assembly Mm_Celera 232000009784844, whole genome shotgun
sequence
Length=8237495

Features flanking this part of subject sequence:
323432 bp at 5' side: uncharacterized protein LOC211208
337262 bp at 3' side: uncharacterized protein LOC73934

Score = 93.5 bits (50), Expect = 2e-17
Identities = 57/60 (95%), Gaps = 1/60 (2%)
Strand=Plus/Plus

Query 1 TACC-CTGTAGGGTTGATAAGCTTATGTTCACTATAACAATTAACACATTTGCCATTGAC 59
|||| ||||| | |||||||||||||||||||||||||||||||||||||||||||||||
Sbjct 577313 TACCTCTGTAAGATTGATAAGCTTATGTTCACTATAACAATTAACACATTTGCCATTGAC 577372
I am wondering whether the ASN file would contain such information (it seems not human readable?). If not, the only way I can think of is to extract chr no. from the standard report by perl or grep (another problem is I don't know how to write perl scripts)... Thanks a lot!

[Growlywolf]

use the module Bio::SearchIO of Bioperl

http://bioperl.open-bio.org/wiki/HOWTO:SearchIO

[logicthief]

use the module Bio::SearchIO of Bioperl

http://bioperl.open-bio.org/wiki/HOWTO:SearchIO

Thanks, growlywolf. It seems very powerful (although not very straightforward for my purpose), I will try it later.