3 <title>JNet Secondary Structure Prediction</title>
6 <strong>JNet Secondary Structure Prediction</strong>
7 <p>Secondary structure prediction methods attempts to infer the
8 likely secondary structure for a protein based on its amino acid
9 composition and similarity to sequences with known secondary structure.
10 The JNet method uses several different neural networks and decides on
11 the most likely prediction via a jury network. <br>
13 <li>Cuff J. A and Barton G.J (1999) Application of enhanced
14 multiple sequence alignment profiles to improve protein secondary
15 structure prediction <em>Proteins</em> <strong>40</strong> 502-511</li>
18 The function available from the
19 <strong>Web Service→Secondary Structure
20 Prediction→JNet Secondary Structure Prediction</strong>
21 menu does two different kinds of prediction, dependent upon the
22 currently selected region:
25 <li>If nothing is selected, and the displayed sequences appear to
26 be aligned, then a JNet prediction will be run for the first sequence
27 in the alignment, using the current alignment. Otherwise the first
28 sequence will be submitted for prediction.</li>
29 <li>If just one sequence (or a region on one sequence) has been
30 selected, it will be submitted to the automatic JNet prediction server
31 for homolog detection and prediction.</li>
32 <li>If a set of sequences are selected, and they appear to be
33 aligned, then the alignment will be used for a Jnet prediction on the <strong>first</strong>
34 sequence selected in the set (that is, the one nearest the top of the
35 alignment window).</li>
37 <p><strong>Note</strong>: JNet secondary structure prediction is a
38 'non-column-separable' service - predictions are based on the sequence
39 profile of contiguous stretches of amino-acid sequence. A prediction
40 will only be made on the visible parts of a sequence (see <a
41 href="../features/hiddenRegions.html">hiding columns</a>) as if it were
42 a contiguous polypeptide chain. Prediction accuracy at the hidden column
43 boundaries may therefore be less than indicated by JNet's own
44 reliability score (see below).</p>
45 <p>The result of a JNet prediction for a sequence is a new annotated
47 <img src="jnetprediction.gif">
48 <p>The sequence for which the prediction was made is the first one
49 in the alignment. If a sequence based prediction was made then the
50 remaining sequences in the alignment are the aligned parts of homologs
51 which were used to construct a sequence profile for the prediction. If
52 the prediction was made using a multiple alignment, then the original
53 multiple alignment will be returned, annotated with the prediction.</p>
54 The annotation bars below the alignment are as follows:
57 <li>Lupas_21, Lupas_14, Lupas_28<br>
58 <em>Coiled-coil predictions for the sequence. These are binary
59 predictions for each location.</em></li>
60 <li>JNETSOL25,JNETSOL5,JNETSOL0<br>
61 <em>Solvent accessibility predictions - binary predictions of 25%,
62 5% or 0% solvent accessibility.</em></li>
64 <em>The consensus prediction - helices are marked as red tubes,
65 and sheets as dark green arrows.</em></li>
67 <em>The confidence estimate for the prediction. High values mean
68 high confidence. prediction - helices are marked as red tubes, and
69 sheets as dark green arrows.</em></li>
71 <em>Alignment based prediction - helices are marked as red tubes,
72 and sheets as dark green arrows.</em></li>
74 <em>HMM profile based prediction - helices are marked as red
75 tubes, and sheets as dark green arrows.</em></li>
77 <em>Jpred prediction - helices are marked as red tubes, and sheets
78 as dark green arrows.</em></li>
80 <em>PSSM based prediction - helices are marked as red tubes, and
81 sheets as dark green arrows.</em></li>
83 <em>Amino Acid frequency based prediction - helices are marked as
84 red tubes, and sheets as dark green arrows.</em></li>
86 <em>A '*' in this annotation indicates that the JNETJURY was
87 invoked to rationalise significantly different primary predictions.</em></li>