Method to construct protein structures

Abstract

A computer-implemented method for creating all-atom, real-space protein conformers. The method involves constructing a backbone structure of α-carbons of a protein from the amino acid sequence of the protein by adding and removing carbon atoms through chain elongation and backtracking. An atom is positioned based on a predicted two-dimensional space, and backtracking removes an atom if it is closer to its neighbour than allowed by van der Waals radii. The method also involves positioning β carbons, C, N, and O atoms to provide favourable bond lengths and bond angles, and positioning sidechain rotamers.

We claim:  
     
         1 . A computer-implemented method for identifying a conformation of a protein of known or unknown structure which comprises the steps of; 
 (a) providing an amino acid sequence of the protein;    (b) constructing a backbone structure of α-carbons of the protein by adding and removing carbon atoms through chain elongation and backtracking, wherein an atom is positioned based on a predicted two-dimensional space, and wherein backtracking removes an atom if it is closer to its neighbor than allowed by van der Waals radii;    (c) positioning β carbons, C, N, and O atoms to provide favourable bond lengths and bond angles; and    (d) positioning sidechain rotamers; thereby outputting a conformation of the protein.    
     
     
         2 . A method as claimed in  claim 1  wherein the conformation of the protein is an all atom protein structure including hydrogen atoms.  
     
     
         3 . A method as claimed in  claim 1  wherein the conformation of the protein is constructed in O(NlogN) time.  
     
     
         4 . A method as claimed in  claim 1  where the conformation of the protein is constructed in real open space.  
     
     
         5 . A method as claimed in  claim 1  which further comprises assembling different conformations of the protein to provide an ensemble of conformations of the protein.  
     
     
         6 . A method as claimed in  claim 5  wherein the ensemble of conformations of the protein are incorporated in a database.  
     
     
         7 . A method as claimed in  claim 6  wherein the database comprises about 50,000 to 500,000 different confirmations of the protein.  
     
     
         8 . A method as claimed in  claim 1  wherein the backbone structure of α-carbons of the protein is constructed by randomly sampling trajectory graphs of amino acid residues representing a statistical sampling for each amino acid residue of the conformational space it is observed to visit in known proteins, wherein the trajectory graphs are resolved into α, β, and coil secondary structure components for each amino acid residue.  
     
     
         9 . A method as claimed in  claim 8  wherein the trajectory graphs are recombined in predicted α, β, and coil secondary structure proportions for each amino acid to form a starting backbone conformation graph.  
     
     
         10 . A method as claimed in  claim 1  wherein in step (b) backtracking is carried out using a hierarchical data structure.  
     
     
         11 . A part of a computer system for identifying a conformation of a protein of known or unknown structure from an amino acid sequence of the protein comprising (a) means for constructing a backbone structure of α-carbons of the protein by adding and removing carbon atoms through chain elongation and backtracking, wherein an atom is positioned based on a predicted two-dimensional space, and wherein backtracking removes an atom if it is closer to its neighbor than allowed by van der Waals radii; (b) means for positioning β carbons, C, N, and O atoms to provide favourable bond lengths and bond angles; and (c) means for positioning sidechain rotamers.  
     
     
         12 . A part of a computer system as claimed in  claim 11  wherein the means for constructing the backbone structure of α-carbons of the protein comprises a trajectory file which defines trajectory graphs of amino acid residues representing a statistical sampling for each amino acid residue of the conformational space it is observed to visit in known proteins, wherein the trajectory graphs are resolved into α, β, and coil secondary structure components for each amino acid residue.  
     
     
         13 . A part of a computer system as claimed in  claim 12  wherein the trajectory graphs are recombined in predicted α, β, and coil secondary structure proportions for each amino acid to form a starting backbone conformation graph.  
     
     
         14 . A part of a computer system as claimed in  claim 11  wherein backtracking is carried out using a hierarchical data structure.  
     
     
         15 . A part of a computer system for identifying favorable areas of conformational space in an ensemble of conformations of a protein comprising (a) a conformer generator module having an input for receiving an amino acid sequence of the protein, means for defining an ensemble of conformations of the protein, and means for recording the amino acid sequence of the protein as an ensemble of conformations of the protein wherein each conformer of the protein is represented by a backbone conformation graph; and (b) a structure analysis module connected to the output of the conformer generator module comprising means for defining a next ensemble of conformers of the protein using a weighting scheme and scoring function; and means for repeating these steps until the backbone conformation graph maintains its shape.  
     
     
         16 . A computer system for constructing backbone structures of α-carbons of a protein comprising a trajectory file which defines trajectory graphs or distributions of amino acid residues representing a statistical sampling for each amino acid residue of the conformational space it is observed to visit in known proteins, wherein the trajectory graphs are resolved into α, β, and coil secondary structure components for each amino acid type; and optionally the graphs are recombined in predicted α, β, and coil secondary structure proportions for each amino acid to form a starting backbone conformation graph.  
     
     
         17 . A computer-implemented process for identifying favorable areas of conformational space in an ensemble of conformations of a protein comprising: 
 (a) providing a database of conformations of the protein wherein each conformer is represented by a backbone conformation graph;    (b) creating a next ensemble of conformations of the protein using a weighting scheme and scoring function; and    (c) repeating step (b) until the backbone conformation graph maintains its shape.    
     
     
         18 . An ensemble of conformations of a protein identified using a method as claimed in  claim 5 .  
     
     
         19 . A database comprising an ensemble of conformations of a protein identified using a method as claimed in  claim 5.