Svar: Re: Svar: Re: [Bioperl-l] tblastn, retrieving no match Aa from subject string

Anders Stegmann anst at
Wed Dec 28 15:01:55 EST 2005

Thanks for replying Jason!

I simply tblastn a protein seq against a chromosome and parse the blast result. the program I am making is supposed to tell me if there is a mutation or not that results in an Aa substitution.

When testing the program I tblastn an orf translated protein seq (which gene is found on chr4) against a downloaded and formated chr4 seq with standaloneblast. the program then tells me that there is a 100% identity between the protein seq and chromosome encoding protein seq.

then I made a Aa substitution (E->Y) at position 2 in the protein seq.
when I tblastn this protein seq against the chr4 seq I get a number like: 349793
because of the: 

foreach ($hsp_obj->seq_inds('hit', 'nomatch')) {

		print $_, "\n";



by 3*349793 I get the position on chr4 that (nearly) corresponds to the start of the codon that encodes the substituted Aa. 

BUT, when I download and format only the gene seq (3750 nt) of the protein (1250 Aa) and use it as a database to which the protein is tblastn against, the scipt gives me the number 2.
Thus, now the scipt gives the Aa position of the substitution instead of the nucleotide codon position.
(it also returns the right Aa (E)that corresponds to the position in the protein)

this is strange!


Anders Stegmann
Ph.d. student
Royal Veterinary and Agricultural University
Institute of Food Science 
Section of Food Microbiology
Rolighedsvej 30
Building 2-74
Room R074
DK-1958 Frederiksberg C
Tlf. +45 35 28 31 58

>>> Jason Stajich <jason.stajich at> 28-12-05 19:17 >>>

On Dec 28, 2005, at 1:08 PM, Anders Stegmann wrote:

> Thanks again, it helped!
> I have two related questions:
> 1) how can I tell if my translated protein is a reverse complement  
> derived from a gene on the opposite chromosome strand?
$hsp->hit->strand  or $hsp->strand('hit)
> 2) why do I get the position and type of the substituted Aa when I  
> blast the protein against its own gene seq, when I only get a  
> position when it is blasted against a chromosome seq? (just out of  
> curiousity)
I don't understand what you mean at all, concrete data example would  
be more helpful.
> Anders.
>>>> Brian Osborne <osborne1 at> 27-12-05 23:52 >>>
> Anders,
> First, think about 3 * 349793...
> Brian O.
> On 12/27/05 5:15 PM, "Anders Stegmann" <anst at> wrote:
>> I get 349793 which I can't explain.
>> what is this number telling me?
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at

Jason Stajich
Duke University

Bioperl-l mailing list
Bioperl-l at

More information about the Bioperl-l mailing list