help/html/webServices/proteinDisorder.html

   1 <html>
   2 <!--
   3  * Jalview - A Sequence Alignment Editor and Viewer (Version 2.8)
   4  * Copyright (C) 2012 J Procter, AM Waterhouse, LM Lui, J Engelhardt, G Barton, M Clamp, S Searle
   5  *
   6  * This file is part of Jalview.
   7  *
   8  * Jalview is free software: you can redistribute it and/or
   9  * modify it under the terms of the GNU General Public License
  10  * as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
  11  *
  12  * Jalview is distributed in the hope that it will be useful, but
  13  * WITHOUT ANY WARRANTY; without even the implied warranty
  14  * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
  15  * PURPOSE.  See the GNU General Public License for more details.
  16  *
  17  * You should have received a copy of the GNU General Public License along with Jalview.  If not, see <http://www.gnu.org/licenses/>.
  18 -->
  19 <head>
  20 <title>JABAWS Protein Disorder Prediction Services</title>
  21 </head>
  22 <body>
  23         <p>
  24                 <strong>JABAWS Protein Disorder Prediction Services</strong> <br />
  25                 The <strong>Web Services&rarr;Disorder</strong> menu in the alignment
  26                 window allows access to protein disorder prediction services provided
  27                 by the configured <a href="http://www.compbio.dundee.ac.uk/jabaws">JABAWS
  28                         servers</a>. Each service operates on sequences in the alignment to
  29                 identify regions likely to be unstructured or flexible, or
  30                 alternately, fold to form globular domains.
  31         </p>
  32         <p>
  33                 Predictor results include both <a href="../features/seqfeatures.html">sequence
  34                         features</a> and sequence associated <a
  35                         href="../features/annotation.html">alignment annotation</a> rows.
  36                 Features display is controlled from the <a
  37                         href="../features/featureSettings.html">Feature Settings</a> dialog
  38                 box. Clicking on the ID for a disorder prediction annotation row will
  39                 highlight or select (if double clicked) the associated sequence for
  40                 that row. You can also use the <em>Sequence Associated</em> option in
  41                 the <a href="../colourSchemes/annotationColouring.html">Colour By
  42                         Annotation</a> dialog box to colour sequences according to the results of
  43                 predictors shown as annotation rows.
  44         </p>
  45         <p>JABAWS 2.0 provides four disorder predictors which are described
  46                 below:</p>
  47         <ul>
  48                 <li><a href="#disembl">DisEMBL</a></li>
  49                 <li><a href="#iupred">IUPred</a></li>
  50                 <li><a href="#ronn">RONN</a></li>
  51                 <li><a href="#globplot">GlobPlot</a></li>
  52         </ul>
  53         <p>
  54                 <strong><a name="disembl"></a><a href="http://dis.embl.de/">DisEMBL
  55                                 (Linding et al., 2003)</a> </strong> <br /> DisEMBL is a set of machine-learning
  56                 based predictors trained to recognise disorder-related annotation
  57                 found on PDB structures.
  58         </p>
  59         <table border="1">
  60                 <tr>
  61                         <td><strong>Name</strong></td>
  62                         <td><strong>Annotation type</strong></td>
  63                         <td><strong>Description</strong></td>
  64                 </tr>
  65                 <tr>
  66                         <td><strong>COILS</strong></td>
  67                         <td>Sequence Feature &amp;<br />Annotation Row
  68                         </td>
  69                         <td>Predicts loops/coils according to DSSP definition<a
  70                                 href="#dsspstates">[1]</a>.<br />Features mark range(s) of residues
  71                                 predicted as loops/coils, and annotation row gives raw value for
  72                                 each residue. Value over 0.516 indicates loop/coil.
  73                         </td>
  74                 </tr>
  75                 <tr>
  76                         <td><strong>HOTLOOPS</strong></td>
  77                         <td>Sequence Feature &amp;<br />Annotation Row
  78                         </td>
  79                         <td>&quot;Hot loops constitute a refined subset of <strong>COILS</strong>,
  80                                 namely those loops with a high degree of mobility as determined from
  81                                 C&alpha; temperature factors (B factors). It follows that highly
  82                                 dynamic loops should be considered protein disorder.&quot;<br />
  83                                 Features mark range(s) of residues predicted to be hot loops and
  84                                 annotation row gives raw value for each residue. Values over 0.6
  85                                 indicates hot loop.
  86                         </td>
  87                 </tr>
  88                 <tr>
  89                         <td><strong>REMARK465</strong></td>
  90                         <td>Sequence Feature &amp;<br />Annotation Row
  91                         </td>
  92                         <td>&quot;Missing coordinates in X-ray structure as defined by
  93                                 remark465 entries in PDB. Nonassigned electron densities most often
  94                                 reflect intrinsic disorder, and have been used early on in disorder
  95                                 prediction.&quot;<br /> Features gives range(s) of residues
  96                                 predicted as disordered, and annotation row gives raw value for each
  97                                 residue. Value over 0.1204 indicates disorder.
  98                         </td>
  99                 </tr>
 100         </table>
 101
 102         <p>
 103                 <a name="dsspstates"></a>[1]. DSSP Classification: &alpha;-helix (H),
 104                 310-helix (G), &beta;-strand (E) are ordered, and all other states
 105                 (&beta;-bridge (B), &beta;-turn (T), bend (S), &pi;-helix (I), and
 106                 coil (C)) considered loops or coils.
 107         </p>
 108
 109
 110         <p>
 111                 <strong><a name="ronn"></a><a
 112                         href="http://www.strubi.ox.ac.uk/RONN">RONN</a></strong> <em>a.k.a.</em>
 113                 Regional Order Neural Network<br />This predictor employs an approach
 114                 known as the 'bio-basis' method to predict regions of disorder in
 115                 sequences based on their local similarity with a gold-standard set of
 116                 disordered protein sequences. It yields a set of disorder prediction
 117                 scores, which are shown as sequence annotation below the alignment.
 118         </p>
 119         <table border="1">
 120                 <tr>
 121                         <td><strong>Name</strong></td>
 122                         <td><strong>Annotation type</strong></td>
 123                         <td><strong>Description</strong></td>
 124                 </tr>
 125                 <tr>
 126                         <td><strong>JRonn</strong>[2]</td>
 127                         <td>Annotation Row</td>
 128                         <td>RONN score for each residue in the sequence. Scores above
 129                                 0.5 identify regions of the protein likely to be disordered.</td>
 130                 </tr>
 131         </table>
 132         <p>
 133                 <em>[2]. JRonn denotes the score for this server because JABAWS
 134                         runs a Java port of RONN developed by Peter Troshin and distributed
 135                         as part of <a href="http://www.biojava.org/">Biojava 3</a>
 136                 </em>
 137         </p>
 138         <p>
 139                 <strong><a name="iupred"></a><a
 140                         href="http://iupred.enzim.hu/Help.php">IUPred</a></strong><br /> IUPred
 141                 employs an empirical model to estimate likely regions of disorder.
 142                 There are three different prediction types offered, each using
 143                 different parameters optimized for slightly different applications. It
 144                 provides raw scores based on two models for predicting regions of
 145                 'long disorder' and 'short disorder'. A third predictor identifies
 146                 regions likely to form structured domains.
 147         </p>
 148         <table border="1">
 149                 <tr>
 150                         <td><strong>Name</strong></td>
 151                         <td><strong>Annotation type</strong></td>
 152                         <td><strong>Description</strong></td>
 153                 </tr>
 154                 <tr>
 155                         <td><strong>Long disorder</strong></td>
 156                         <td>Annotation Row</td>
 157                         <td>Prediction of context-independent global disorder that
 158                                 encompasses at least 30 consecutive residues of predicted disorder.
 159                                 Employs a 100 residue window for calculation.<br />Values above 0.5
 160                                 indicates the residue is intrinsically disordered.
 161                         </td>
 162                 </tr>
 163                 <tr>
 164                         <td><strong>Short disorder</strong></td>
 165                         <td>Annotation Row</td>
 166                         <td>Predictor for short, (and probably) context-dependent,
 167                                 disordered regions, such as missing residues in the X-ray structure
 168                                 of an otherwise globular protein. Employs a 25 residue window for
 169                                 calculation, and includes adjustment parameter for chain termini
 170                                 which favors disorder prediction at the ends.<br />Values above 0.5
 171                                 indicate short-range disorder.
 172                         </td>
 173                 </tr>
 174                 <tr>
 175                         <td><strong>Structured domains</strong></td>
 176                         <td>Sequence Feature</td>
 177                         <td>Features highlighting likely globular domains useful for
 178                                 structure genomics investigation. <br />Post-analysis of disordered
 179                                 region profile to find continuous regions confidently predicted to
 180                                 be ordered. Neighbouring regions close to each other are merged,
 181                                 while regions shorter than the minimal domain size of at least 30
 182                                 residues are ignored.
 183                         </td>
 184                 </tr>
 185         </table>
 186         <p>
 187                 <strong><a name="globplot"></a><a
 188                         href="http://globplot.embl.de/">GLOBPLOT</a></strong><br /> Defines regions
 189                 of globularity or natively unstructured regions based on a running sum
 190                 of the propensity of residues to be structured or unstructured. The
 191                 propensity is calculated based on the probability of each amino acid
 192                 being observed within well defined regions of secondary structure or
 193                 within regions of random coil. The initial signal is smoothed with a
 194                 Savitzky-Golay filter, and its first order derivative computed.
 195                 Residues for which the first order derivative is positive are
 196                 designated as natively unstructured, whereas those with negative
 197                 values are structured.<br />
 198         <table border="1">
 199                 <tr>
 200                         <td><strong>Name</strong></td>
 201                         <td><strong>Annotation type</strong></td>
 202                         <td><strong>Description</strong></td>
 203                 </tr>
 204                 <tr>
 205                         <td><strong>Disordered Region</strong></td>
 206                         <td>Sequence Feature</td>
 207                         <td><br />Sequence features marking range(s) of residues with
 208                                 positive dydx values (correspond to the #Disorder column from JABAWS
 209                                 results)</td>
 210                 </tr>
 211                 <tr>
 212                         <td><strong>Globular Domain</strong>
 213                         <td>Sequence Feature</td>
 214                         <td>Putative globular domains</td>
 215                 </tr>
 216                 <tr>
 217                         <td><strong>Dydx</strong></td>
 218                         <td>Annotation row</td>
 219                         <td>First order derivative of smoothed score. Values above 0
 220                                 indicates residue is disordered.</td>
 221                 </tr>
 222                 <tr>
 223                         <td><strong>Smoothed Score<br />Raw Score
 224                         </strong></td>
 225                         <td>Annotation Row</td>
 226                         <td>The smoothed and raw scores used to create the differential
 227                                 signal that indicates the presence of unstructured regions.<br /> <em>These
 228                                         are hidden by default, but can be shown by right-clicking on the
 229                                         alignment annotation panel and selecting <strong>Show
 230                                                 hidden annotation</strong>
 231                         </em>
 232                         </td>
 233                 </tr>
 234         </table>
 235         <p>
 236                 <em>Documentation and thresholds for the JABAWS Disorder
 237                         predictors adapted from a personal communication by Nancy Giang,
 238                         2012.</em>
 239         </p>
 240 </body>
 241 </html>