Iam a student, just started working with protein mutations and I used NCBI info about different domains present in ITK. I used NCBI because it had much more data e.g. PI binding sites. However, when I aligned those domains with my mutations, I was curious and checked those domain sequences with Uniprot. As presented below, there are some discrepancies between NCBI and Uniprot (I checked, it's the same Homo sapiens ITK). I wanted to ask if this is a common thing, that such data differs and which source is more reliable.
PH Domain according to Uniprot and NCBI:
http://i.imgur.com/j8BNqWl.jpg
Iam aware that the amino acids at the beginning and the end of a domain are not as important as the ones in the middle, especially when special binding sites need to be formed, like in this case. But saying that a mutation is within some domain or not, or even that given domain is 108 or 104, amino acids length seems inconsistent with different data from the online databases. If this is normal, how do you deal with it?
Links:
NCBI http://www.ncbi.nlm....r/?id=NM_005546
Uni http://www.uniprot.org/uniprot/Q08881
i agree with Nicolas..Uniport gives u a detail information.. beyond that u also check ur domin inforamtion in pfam server. it is also provides u a bunch of annotations on ur protein domain
http://pfam.sanger.ac.uk/
Dear Filip
There are several protein data banks (PDBs) include UNIPROT, SWISSPROT and ... . In using of them, it is important to identify your goals clearly. If you want to work with your protein's mRNA and DNA, NCBI is useful but for comparing the primary to tertiary structure of proteins between species you must use one of the mentioned PDBs.
However, you must know that information about certain protein are similar between different PDBs but the type of presentation is different.
Cheers
If you want to test domains in ncbi, you might select Conserved Domains Database search option (http://www.ncbi.nlm.nih.gov/Structure/cdd/docs/cdd_search.html) by using protein sequence.
You can also use pfam (pfam.sanger.ac.uk/) or ProDom (prodom.prabi.fr/) or Interpro (www.ebi.ac.uk/interpro/) databases for domain search.
Always the results showed links to uniprot because, uniprot is the universal protein sequence database.
So, you can rely even the results of NCBI, CDD database.
- Badges
- Science topic
- Similar topics
- Chemistry
- Biochemistry