1. Introduction
2. Authors and publications
3. Organization of PCBOST
4. The guide to the system

PCBOST is the Protein Classification Based On Structural Trees. This structural classification of proteins is only based on the spatial structural similarity and common folding pathways simulated with the trees. The classification disregards the amino acid sequences, functions, and evolutionary relationships of proteins which are considered, to some extent, in other known classifications. The structural tree of proteins is a set of all allowable intermediate and final 3D structures that can be derived from the root (starting) structure by stepwise addition of α-helices and/or β-strands to the growing structure taking into account a restricted set of rules inferred from the known principles of protein structure. A structural motif having a unique polypeptide chain fold is used as a root structure of a tree. Possible pathways of structural complications are shown with lines, which eventually integrate all structures into one tree.

In the classification, following levels of hierarchy are used: Class – Structural Tree – Level – Fold – Protein Domain – Species – PDB Entry. All proteins and domains belonging to one tree can be assigned to one Structural Tree. All proteins and domains belonging to different levels of one Structural Tree form a list of Levels. All proteins and domains from one Level that have the same arrangement of secondary elements form Folds.

Authors: A.B. Gordeev, E.A. Boshkova, A.M. Kargatov, A.V. Efimov.

Protein Structure Modeling Group, Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow Region, 142290 Russia.

Publications:

1. Gordeev A.B., Efimov A.V. (2012) Modeling of folds and folding pathways for some protein families of (α+β)- and α/β-classes. J. Biomol. Struct. Dyn., DOI: 10.1080/07391102.2012.691341.
2. Gordeev A.B., Kargatov A.M., Efimov A.V. (2010) PCBOST: Protein classification based on structural trees. Biochemical and Biophysical Research Communications, 397, 470-471.
3. Kargatov A.M., Efimov A.V. (2010) A novel structural motif and structural trees for proteins containing it. Biochemistry (Moscow), 75, 249-256.
4. Gordeev A.B., Efimov A.V. (2009) Novel structural tree of (α+β)-proteins containing abCd units. Mol. Biol. (Moscow), 43, 480-484.
5. Gordeev A.B., Kondratova M.S., Efimov A.V. (2008) Novel structural tree of β-proteins containing abcd units. Mol. Biol. (Moscow), 42, 285-288.

If you have got any notes and wishes you may use our E-mail:
      gordeew@vega.protres.ru.

At present the classification database for eighteen structural classes of proteins has been developed. The database includes: β-proteins containing abcd-Units and three-β-corners; (α+β)-proteins containing abCd-, a₁abCd-, a₂a₁abCd-Units and Coiled β-hairpins; α/β-proteins containing β-α-β-ψ-motifs, ψ-β-α-β-motifs, five-segment, seven-segment and eight-segment α/β-motifs, analogs of the seven-segment α/β-motif having S-like β-sheet and 6 structural classes of β- and (α+β)-proteins with motifs containing an S-like β-sheet.

At the schematic representations of structural trees, β-strands are shown as squares and α-helices as circles. The linkers directed toward and away from the viewer are shown with double and single lines, respectively. The transfers of an element from one layer to other layer are shown with dotted lines. The grow of the protein folds is realized by stepwise addition of secondary structural elements to the corresponding structural motifs.

1. The structural tree of β-proteins containing abcd-Units includes 528 proteins (among them 244 are nonhomologous; total 1,511 PDB entries). The tree is composed of 7 Levels and proteins are grouped into 81 Folds.
2. The structural tree of β-proteins containing tree-β-corners includes 573 proteins (among them 204 are nonhomologous; total 2,570 PDB entries). The tree is composed of 10 Levels and proteins are grouped into 78 Folds.
3. The structural tree of (α+β)-proteins containing abCd-Units includes 926 proteins (among them 401 are nonhomologous; total 2,636 PDB entries). There are 10 Levels in the tree and proteins are grouped into 98 Folds.
4. The structural tree of (α+β)-proteins containing a₁abCd-Units includes 283 proteins (among them 104 are nonhomologous; total 1,474 PDB entries). There are 11 Levels in the tree and proteins are grouped into 39 Folds.
5. The structural tree of (α+β)-proteins containing a₂a₁abCd-Units includes 84 proteins (among them 52 are nonhomologous; total 275 PDB entries). There are 8 Levels in the tree and proteins are grouped into 24 Folds.
6. The structural tree of (α+β)-proteins containing coiled beta-hairpins includes 313 proteins (among them 136 are nonhomologous; total 900 PDB entries). There are 7 Levels in the tree and proteins are grouped into 49 Folds.
7. Six structural trees of β- and (α+β)-proteins with motifs containing an S-like β-sheet include 401 proteins (total 1517 PDB entries). There are 31 Levels in the trees and proteins are grouped into 56 Folds.
8. The structural tree of α/β-proteins containing five-segment α/β-motifs includes 1,301 proteins (among them 388 are nonhomologous; total 3,926 PDB entries). There are 16 Levels in the tree and proteins are grouped into 172 Folds.
9. The structural tree of α/β-proteins containing seven-segment α/β-motifs includes 870 proteins (among them 294 nonhomologous; total 2,567 PDB entries). There are 12 Levels in the tree and proteins are grouped into 85 Folds.
10. The structural tree of α/β-proteins containing eight-segment α/β-motifs includes 34 proteins (total 80 PDB entries). There are 6 Levels in the tree and proteins are grouped into 10 Folds.
11. The structural tree of α/β-proteins containing analogs of the seven-segment α/β-motif having the S-like β-sheet includes 90 proteins (total 258 PDB entries). There are 6 Levels in the tree and proteins are grouped into 10 Folds.
12. The structural tree of α/β-proteins containing β-α-β-ψ-motifs includes 218 proteins (among them 77 nonhomologous; total 655 PDB entries). There are 7 Levels in the tree and proteins are grouped into 24 Folds.
13. The structural tree of α/β-proteins containing ψ-β-α-β-motifs includes 167 proteins (among them 58 nonhomologous; total 668 PDB entries). There are 11 Levels in the tree and proteins are grouped into 25 Folds.

This site is organized with an application of xml and php technologies. If your browser does not keep xml technology, you need to install a recent version of the browser.

1. Put the button “HELP” to go to the help page.
   
   
2. Put the button “Main tree” to see the whole structural tree. You can see all the Folds that form the tree. Folds are numbered left to right. The Folds actually found in proteins are framed. You may look their content in the main classification system.
   
   
3. Click on the image with the drawn Fold to look a list of found proteins and domains that belong to this Fold.
   
   
4. Being in the main classification system you may always go to the structural tree, at that part of the tree where the Fold is located. So you need to put the link “Structural Tree”, which is located on the right of the image with the drawn Fold.
   
   
5. The Structural tree consists of two parts. First, you may use the left part of the tree. Put the pointer under the inscription “Go to another part of the tree” to look the right part of the tree. Put the analogous pointer to go to the left part.
   
   
6. Put the link “Go To Begin” to go to begin.
   
   
7. The link “β-proteins with abcd unit” enters the list of links “Levels” pulling down to one of the hierarchical levels of the system. Here, the list of Levels is shown.
   
   
8. Put one of the links “Level” to go to another level of the hierarchical levels of the system. Here is the list of proteins or domains.
   
   
9. If you put one of the protein links across the level “Species” you may get the list of links with PDB-codes. You have got an opportunity to make a copy of the PDB file. If a protein consists of some domains or subunits, you can see a sign of the subunits and/or numbers of amino acid residues
   
   
10. Put one of the links at the list shown on the left up of every level to return to previous levels. In this list the links are arranged in the order from the higher level of hierarchy to lower one.
    
    
11. If you find out a localization of a protein in the tree, you have got an opportunity to find it at the system. There are two opportunities of the search: the search of a protein by homology and the search by analogy.
    
    
    • The search of a protein based on homology. If a protein contains one of the root structures, and its PDB-code is known you may locate it at the tree. So you need:
      
      
      1. put the link “Fold” and go to the list of found proteins with analogous structure.
         
         
      2. put the link “Structural Tree” and look a localization of the protein in the tree.
         
      
      
    • The search of a protein based on analogy. If the structure of a protein is known but its PDB-code is not known, you may locate it at the tree. So you need:
      
      
      1. find the structure of the protein at the structural tree (see 2-4).
         
         
      2. click on the image with the drawn Fold and go to a list of analogous proteins with the same structure
         
         
      3. choose the protein from the list.
      
      
    • If you can not find a protein, it is not included in the system yet.
    
    
12. Searching system for PDB-codes.
    
    
    • Put the button “Help”.
      
      
    • It is necessary to locate the PDB-code at the window under the inscription “Enter PDB-code”.
      
      
    • Put the button “Search…”.
      
      
    • Put one of the results and the new page with the protein will appear. Then close the searching window and open this page.
      
      
    • If you can see the inscription “No Found” at the searching window, the system has not included this protein yet.
      
      
    • If you can see the inscription “You put no PDB-code”, you have really put no PDB-code at the searching window.