<!------------Description---------->
This track shows the positions and names of predicted domains within the 
protein as 
determined by Hidden Markov Modeling (<A HREF="http://www.cse.ucsc.edu/research/compbio/sam.html" 
TARGET=_blank>HMM</A>) comparison to all known 3-D structures. Each predicted
domain is represented by a yellow box. Click on a domain to display information 
about that domain on the 
<A HREF="http://supfam.mrc-lmb.cam.ac.uk/SUPERFAMILY/"
TARGET=_blank>Superfamily</A> website. 
SuperFamily in turn relies on the Structural Classification of
Proteins System (<A HREF="http://scop.mrc-lmb.cam.ac.uk/scop/" 
TARGET=_blank>SCOP</A>). This method has an advantage over Blast or
psi-Blast in that protein folds are often conserved long after primary
sequence alignment has become unreliable (10-25% identity range).
<P>
InterPro domains, based on clustering of sequence alignments, do not
require a  3-D reference sequence and therefore complement SuperFamily domains.
Refer to the <A HREF="pbintPCnt.shtml">InterPro Domains histogram</A> to view 
the total number of predicted domains within a protein and compare this 
count to the number of multiple-domain proteins genome-wide.
