Login

Join for Free!
117057 members


Patterns and profiles.

Everything on bioinformatics, the science of information technology as applied to biological research.

Moderator: BioTeam

Patterns and profiles.

Postby Morris » Sat Oct 22, 2005 2:52 pm

hi at all.Someone can tell me the difference between pattern and prfile and between global and local allinment? Thank you.
User avatar
Morris
Death Adder
Death Adder
 
Posts: 57
Joined: Sat Oct 22, 2005 2:42 pm
Location: Italy

Postby Juanu » Tue Nov 08, 2005 6:01 pm

Hi!,

For what I understand, the difference between a global and local alignment, is that the global alignment considers all the letters from the sequence to do the alignment. So, in two sequences of 100 characters the global alignment will try to put align each other in the same order (1 with 1, 2 with 2, etc...).
In a local alignment, the search is done over small regions, so you can find related regions on your sequences. In this case it is possible to align region 10-20 from one sequence with region 50-60 of the other, because the method will allow you to do that search.
In any case, global alignment methods aren´t that used any more, because local alignment methods are best to find similarity between sequences (for example between two proteins of different organisms), allowing more flexibility. An example of a very popular local alignment tool is BLAST ( http://www.ncbi.nlm.nih.gov/BLAST/), where you can put your sequence and look for similar sequences in other organisms.

Regarding the differences between a profile and a pattern, I´ll do my best to explain it (and in english!!...:P ) When you align several sequences together, like an enzyme from different organisms (but that have the same function in all of them), and you look for the most repeated characters on them, you will generate a pattern of that protein that you can use to search for the same function on other proteins that you only have the sequence and no experiment that can give you the function.
The problem is that this only works for very similary proteins, because you will search always for the same characters. To avoid that, you generate a profile of this pattern, that consists in a numerical matrix associated with the different characters. In this way you have a more flexible way to look for your pattern in other sequences, that could be not so close related to the ones that originated your pattern.

I am sorry if I wasn´t very clear with this answer, but I´ll try to improve it. Please let me know any comments!...

Bye!
User avatar
Juanu
Garter
Garter
 
Posts: 5
Joined: Mon Nov 07, 2005 1:13 pm
Location: Santiago, Chile

Postby Morris » Tue Nov 08, 2005 7:44 pm

No, thanks, I've understood.
http://www.forumbiotech.forumfree.net(In italian)
http://www.forumfree.net/?c=37503 (international medicine forum)(In english)

http://u2forum.forumfree.net (in Italian)
User avatar
Morris
Death Adder
Death Adder
 
Posts: 57
Joined: Sat Oct 22, 2005 2:42 pm
Location: Italy


Postby Juanu » Wed Nov 09, 2005 1:09 am

yeah!...I can write and be understood!!!...
jeje...

:D :D :D
User avatar
Juanu
Garter
Garter
 
Posts: 5
Joined: Mon Nov 07, 2005 1:13 pm
Location: Santiago, Chile


Return to Bioinformatics

Who is online

Users browsing this forum: No registered users and 0 guests

cron