xSpecifyQuery

 xSpecifyQuery returns the description of a model motif

 The variable Query has several fields, most of which are optional:
   Query.Description    a useful string, can be long
   Query.Name           a short string, which will become part of a filename

 For all searches, these are required:
   Query.SearchFiles    a cell array of PDB filenames or PDB lists

 For geometric searches, these are required:
   Query.Filename       a string like 1s72, where the query motif is found
   Query.NTList         a list of nucleotide numbers
   Query.ChainList      a list of chain specifications, use if needed

 For non-geometric or mixed searches, these are optional:
   Query.Edges          list of required basepairing or stacking interactions
                        also specify allowed or disallowed pairs here
   Query.Mask           a mask for which nucleotides to allow (defaults N)
   Query.AngleWeight    weights to put on the angles (defaults 1)
   Query.DistanceWeight weights to put on nucleotide distances (defaults 1)
   Query.DiscCutoff     discrepancy cutoff D_0 (default 0.4)
   Query.RelCutoff      relaxed cutoff D_1 (default Query.DiscCutoff)
   Query.MaxDiff        maximum difference between sorted nucleotide indices
   Query.MinDiff        minimum difference between sorted nucleotide indices

   Query.Geometric      set to 0 to ignore geometry, only use screens. 
                        Default is 1.
   Query.ExcludeOverlap set to 1 to eliminate highly redundant motifs with
     larger discrepancies; often, you get the same candidate with one or two
     nucleotides different but much higher discrepancy.  Default is 1 when
     Query.NumNT > 6.

0001 % xSpecifyQuery returns the description of a model motif
0002 %
0003 % The variable Query has several fields, most of which are optional:
0004 %   Query.Description    a useful string, can be long
0005 %   Query.Name           a short string, which will become part of a filename
0006 %
0007 % For all searches, these are required:
0008 %   Query.SearchFiles    a cell array of PDB filenames or PDB lists
0009 %
0010 % For geometric searches, these are required:
0011 %   Query.Filename       a string like 1s72, where the query motif is found
0012 %   Query.NTList         a list of nucleotide numbers
0013 %   Query.ChainList      a list of chain specifications, use if needed
0014 %
0015 % For non-geometric or mixed searches, these are optional:
0016 %   Query.Edges          list of required basepairing or stacking interactions
0017 %                        also specify allowed or disallowed pairs here
0018 %   Query.Mask           a mask for which nucleotides to allow (defaults N)
0019 %   Query.AngleWeight    weights to put on the angles (defaults 1)
0020 %   Query.DistanceWeight weights to put on nucleotide distances (defaults 1)
0021 %   Query.DiscCutoff     discrepancy cutoff D_0 (default 0.4)
0022 %   Query.RelCutoff      relaxed cutoff D_1 (default Query.DiscCutoff)
0023 %   Query.MaxDiff        maximum difference between sorted nucleotide indices
0024 %   Query.MinDiff        minimum difference between sorted nucleotide indices
0025 %
0026 %   Query.Geometric      set to 0 to ignore geometry, only use screens.
0027 %                        Default is 1.
0028 %   Query.ExcludeOverlap set to 1 to eliminate highly redundant motifs with
0029 %     larger discrepancies; often, you get the same candidate with one or two
0030 %     nucleotides different but much higher discrepancy.  Default is 1 when
0031 %     Query.NumNT > 6.
0032 
0033 function [Query] = xSpecifyQuery(QName);
0034 
0035 if nargin > 0,
0036   Query.Name = QName;
0037 else                        % change the following line to change the query!
0038   Query.Name = 'BasepairGeometric2';
0039   Query.Name = 'Sarcin9Mixed';
0040   Query.Name = 'Sarcin5Geo';
0041   Query.Name = 'StackedPair'; 
0042   Query.Name = 'Stack';
0043   Query.Name = 'Basepair';
0044   Query.Name = 'Sarcin5Symb';
0045 end
0046 
0047 Query.SearchFiles = '1s72';        % default is to search 1s72
0048 
0049 switch Query.Name
0050 
0051 % ---------------------------------- Searches from FR3D paper by Sarver et al
0052 
0053 case 'Sarcin5Geo'
0054   Query.Description    = 'Sarcin five nucleotide geometric';
0055   Query.Filename       = '1s72';
0056   Query.NTList         = {'2694' '2701' '2693' '2702' '2692'};
0057   Query.ChainList      = {'0' '0' '0' '0' '0'};   % all in the 23S
0058   Query.DiscCutoff     = 0.5;
0059   Query.SearchFiles    = {'1s72' 'Nonredundant_list'};
0060   Query.SearchFiles    = {'1s72' 'HighResolution_list'};
0061 
0062 case 'Sarcin5Symb'
0063   Query.Description    = 'Sarcin five nucleotide symbolic';
0064   Query.Edges{1,2}     = 'tHS';
0065   Query.Edges{3,5}     = 'cHS';
0066   Query.Edges{3,4}     = 'tWH';
0067   Query.MaxDiff(5,3)   = 2;
0068   Query.MaxDiff(3,1)   = 2;
0069   Query.MaxDiff(4,2)   = 2;
0070 
0071 case 'Sarcin7Mixed'
0072   Query.Description    = 'Sarcin seven nucleotide mixed';
0073   Query.Filename       = '1s72';
0074   Query.NTList         = {'2694' '2701' '2693' '2702' '2692' '2691' '2703'};
0075   Query.ChainList      = {'0' '0' '0' '0' '0' '0' '0'};   % all in the 23S
0076   Query.Edges{3,4}     = 'tWH';
0077   Query.DiscCutoff     = 0.5;       
0078   Query.ExcludeOverlap = 0;
0079 
0080 case 'Sarcin9Mixed'
0081   Query.Description    = 'Sarcin nine nucleotide mixed';
0082   Query.Filename       = '1s72';
0083   Query.NTList         = {'2694' '2701' '2693' '2702' '2692' '2691' '2703' '2690' '2704'};
0084   Query.ChainList      = {'0' '0' '0' '0' '0' '0' '0' '0' '0'};% all in the 23S
0085   Query.Edges{3,4}     = 'tWH';
0086   Query.DiscCutoff     = 0.5;
0087   Query.ExcludeOverlap = 1;
0088   Query.SearchFiles    = {'1s72' 'Nonredundant_list'};
0089 
0090 case 'KinkTurnCentral'
0091   Query.Description    = 'Kink-turn central base mixed';
0092   Query.Filename       = '1s72';
0093   Query.NTList         = {'80' '97' '81' '93' '94' '98'};
0094   Query.ChainList      = {'0' '0' '0' '0' '0' '0'};   % all in the 23S
0095   Query.Edges{1,2}     = 'tHS';
0096   Query.DiscCutoff     = 0.7;  
0097   Query.ExcludeOverlap = 1;
0098 
0099 case 'KinkTurnClosing'
0100   Query.Description    = 'Kink-turn closing base pair mixed';
0101   Query.Filename       = '1s72';
0102   Query.NTList         = {'80' '97' '81' '93' '100' '77'};
0103   Query.ChainList      = {'0' '0' '0' '0' '0' '0'};   % all in the 23S
0104   Query.Edges{1,2}     = 'tHS';
0105   Query.DiscCutoff     = 0.9;    
0106   Query.ExcludeOverlap = 1;
0107 
0108 case 'GNRA4NonSeq'
0109   Query.Description    = 'GRNA hairpin without sequential constraint';
0110   Query.Filename       = '1s72';
0111   Query.NTList         = {'804' '805' '808' '809'};
0112   Query.ChainList      = {'0' '0' '0' '0'}; 
0113   Query.Edges{1,4}     = 'cWW';
0114   Query.Edges{2,3}     = 'tSH';
0115   Query.DiscCutoff     = 1;      
0116   Query.ExcludeOverlap = 1;
0117 
0118 case 'GNRA4Seq'
0119   Query.Description    = 'GRNA hairpin with sequential constraint';
0120   Query.Filename       = '1s72';
0121   Query.NTList         = {'804' '805' '808' '809'};
0122   Query.ChainList      = {'0' '0' '0' '0'}; 
0123   Query.Edges{1,4}     = 'cWW';
0124   Query.Edges{2,3}     = 'tSH';
0125   Query.DiscCutoff     = 1;      
0126   Query.MaxDiff(1,4)   = 6;
0127   Query.ExcludeOverlap = 1;
0128 
0129 case 'GNRA5'
0130   Query.Description    = 'GRNA hairpin 5 nucleotide';
0131   Query.Filename       = '1s72';
0132   Query.NTList         = {'804' '805' '807' '808' '809'};
0133   Query.ChainList      = {'0' '0' '0' '0' '0'}; 
0134   Query.Edges{1,5}     = 'cWW bif';
0135   Query.Edges{1,2}     = 's35';
0136   Query.Edges{3,4}     = 's35';
0137   Query.DiscCutoff     = 0.8;
0138   Query.MaxDiff(1,5)   = 6;
0139   Query.MaxDiff(2,4)   = 4;
0140   Query.ExcludeOverlap = 1;
0141 
0142 % -------------------------------------------- Additional searches
0143 
0144 case 'BasepairGeometric'
0145   Query.Filename   = '1s72';
0146   Query.NTList     = {'804' '809'};
0147   Query.DiscCutoff = 0.4;
0148 
0149 case 'BasepairGeometric2'
0150   Query.Filename   = '1u6b';
0151   Query.NTList     = {'59','85'};
0152   Query.DiscCutoff = 0.5;
0153   Query.SearchFiles    = {'Nonredundant_list'};
0154 
0155 case 'Basepair'
0156 %  Query.Edges{1,2}     = 'nPB';
0157   Query.Edges{1,2}      = 'B2P';
0158 %  Query.Diff{1,2}      = '> <1000';
0159  % Query.Config{1}      = 'syn';
0160  % Query.SearchFiles    = {'1s72', '1j5e'};
0161 
0162 case 'GNRA4NonSeq'
0163   Query.Description    = 'GRNA hairpin without sequential constraint';
0164   Query.Filename       = '1s72';
0165   Query.NTList         = {'804' '809' '805' '808'};
0166   Query.ChainList      = {'0' '0' '0' '0'}; 
0167   Query.Mask           = 'NNNN';
0168   Query.Edges{1,2}     = 'ncWW';
0169   Query.Edges{3,4}     = 'ntSH';
0170   Query.DiscCutoff     = 1;         % guaranteed to find all candidates
0171                                     % with discrepancy below this number
0172   Query.ExcludeOverlap = 1;
0173 
0174 case 'StackedPair'
0175   Query.Description    = 'Stacked pairs';
0176   Query.Diff{1,2}      = '> <5';
0177   Query.Diff{4,3}      = '< <5';
0178   Query.Edges{1,4}     = 'cWW';
0179   Query.Edges{2,3}     = 'cWW';
0180   Query.Edges{1,2}     = 'stack';
0181   Query.Edges{3,4}     = 'stack';
0182   Query.Mask           = 'GGNN';
0183 
0184 case 'Stack'
0185   Query.Description    = 'Two stacked bases';
0186   Query.Mask           = 'AA';
0187   Query.Edges{1,2}     = 'stack';
0188   Query.SearchFiles    = {'1s72'};
0189 
0190 case 'StackedOncWW'
0191   Query.Description    = 'What stacks on a cWW?';
0192   Query.Edges{1,2}     = 'cWW';
0193   Query.Edges{3,4}     = 'Pair ~tSH ~tHS ~cWW';  % exclude categories we know
0194   Query.Edges{1,3}     = 'Stack';
0195   Query.Edges{2,4}     = 'Stack';
0196   Query.Diff{1,3}      = '=1';
0197   Query.Diff{2,4}      = '=1';
0198 
0199 case 'Sarcin5Quick'
0200   Query.Description    = 'Sarcin five nucleotide quick search';
0201   Query.Filename       = '1s72';
0202   Query.NTList         = {'2694' '2701' '2693' '2702' '2692'};
0203   Query.ChainList      = {'0' '0' '0' '0' '0'};   % all in the 23S
0204   Query.DiscCutoff     = 0.3;         % guaranteed to find all candidates
0205                                       % with discrepancy below this number
0206 case 'cWW-noncWW-cWW'
0207   Query.Description    = 'Noncanonical pair between canonical';
0208   Query.Edges{1,6}     = 'cWW';
0209   Query.Edges{2,5}     = '~cWW';
0210   Query.Edges{3,4}     = 'cWW';
0211   Query.MaxDiff(1,2)   = 1;
0212   Query.MaxDiff(2,3)   = 1;
0213   Query.MaxDiff(4,5)   = 1;
0214   Query.MaxDiff(5,6)   = 1;
0215 
0216 end
0217 
0218 % List of nucleotide mask codes:
0219 % A-A
0220 % C-C
0221 % G-G
0222 % U-U
0223 % A,C-M
0224 % A,G-R
0225 % A,U-W
0226 % C,G-S
0227 % C,U-Y
0228 % G,U-K
0229 % A,C,G-V
0230 % A,C,U-H
0231 % A,G,U-D
0232 % C,G,U-B
0233 % A,C,G,U-N
0234