Residue pairs sorted by strength in covariance:
Legend: The darker and larger the blue dots, the higher strength in covariance. Below we provide the list of the top [1.5 x length] gremlin predictions, sequence seperation > 3. The i and j are positions as given in the consensus sequence. Show Scaled Distribution
Legend: The value of the raw score is the function of the learning procedure, L2 normalization and APC (entropic) correction. These are to be used for relative ranking only.
Top (length/2) GREMLIN results overlayed on top 10 PDB hits:
Legend: The grey circles underneath are pdb residue contacts (min distance < 5 Angstroms). The coloring of these circles is based on HHsearch results which uses the overall probability, per-site alignment prob and agreement of top hits weighted by HHsearch score (Note we only consider monomeric contacts, there might be homo-oligomeric contacts in the pdb that are not shown.)
If you are interested in a protein containing this domain,
drop us a note and we'll try to predict the structure for you.