[BiO BB] protein families
pooja at igc.gulbenkian.pt
pooja at igc.gulbenkian.pt
Fri Sep 26 05:57:17 EDT 2003
You can find all the proteins classified in superfamily, at any of the
1. SuperFamily - Classification is based on profile hidden Markove Model
that represents all proteins of known structure based on SCOP (Structural
Classification Of Proteins).
2. PIR SuperFamily (PIRSF) - The classification system is based on
evolutionary relationship of whole proteins.
If you are interested in identifying a possible family for an
uncharacterized protein ( as presently I am looking for) you may be
interested to try InterPro.
Interpro is a joint effort of proteins sequence databases like SWISS-PROT
and TrEMBL, functional sites, motifs and domain databases, PRINTS, Pfam,
and ProDom, and resources for Protein families, like PIRSF and
SUPERFAMILY. Features like functional sites, motifs and domains which
are exptracted from known protein sequences and known protein families are
applied to unknown protein sequence while making the prediction.
But the sequences I am trying with, InterPro always leaves me with the
same family status I have started from...... Unknown protein or
Hypothetical protein !
So I am also looking for a good protein family prediction tool, or Gene
Family Prediction tool. I will be very greatful if someone from the list
can give me some insight
I will soon look into the suggested Bioinformatics readings. May be
helpful for me as well.
> Depends on what you mean by "family"
> CATH and SCOP assign proteins in a hierarchical manner to classes, folds,
> superfamilies, families based on sequence based (SCOP) and structure based
> (CATH) similarities. Both are manually or semi-manually curated. FSSP does
> the same task automatically. All three are in a high rate of agreement
> (75-80%) regarding their calssifications. Of course, you are limited to
> the proteins in PDB only: the ones for which there are solved structures.
> If you wish to consider more sequecnes, then other assignemetns are
> possible, depending on your purpose. Pfam is a good example.
> I suggest you get a good bioinformatics textbook, such as David Mount's,
> or Baxevantis, and look through the different classification schemes, and
> choose the one which suits your purpose best.
> Iddo Friedberg, Ph.D.
> The Burnham Institute
> 10901 N. Torrey Pines Rd.
> La Jolla, CA 92037, USA
> Tel: +1 (858) 646 3100 x3516
> Fax: +1 (858) 646 3171
> On Fri, 26 Sep 2003, Liu Haifeng wrote:
>> I am trying to find a collection of protein sequences which have been
>> correctly assigned to different families. Anybody can suggest where to
>> obtain such kind of data? PDB and Swiss-Prot seem to provide sequences
>> without family information.
>> Would appreciate your help, thanks a lot!
>> Haifeng Liu
>> BiO_Bulletin_Board maillist - BiO_Bulletin_Board at bioinformatics.org
> BiO_Bulletin_Board maillist - BiO_Bulletin_Board at bioinformatics.org
More information about the BBB