+ will be coloured with that colour.</p>
+ <hr/>
+ <p><strong><a name="seqgrprefs">SEQUENCE_REF and GROUP_REF</a></strong><br/>
+ By
+ default, annotation is associated with the alignment as a whole.
+ However, it is also possible to have an annotation row associated with
+ a specific sequence, or a sequence group. Clicking the annotation
+ label for sequence or group associated annotation will highlight the
+ associated rows in the alignment, and double clicking will select
+ those rows, allowing further analysis. While group associated
+ annotation remains associated with a particular alignment, sequence
+ associated annotation can move with a sequence - so copying a sequence
+ to another alignment will also copy its associated annotation.
+ </p>
+ <p>You can associate an annotation with a sequence by preceding its
+definition with the line:
+<pre>SEQUENCE_REF	<em>seq_name</em>	<em>[startIndex]</em></pre>
+All Annotations defined after a SEQUENCE_REF command will then be
+associated with that sequence, and the first field in the Value field
+list will (optionally) be placed at the <em>startIndex</em>'th column.</p>
+
+<p>Sequence associations are turned off for subsequent annotation
+definitions by:
+<pre>SEQUENCE_REF	ALIGNMENT</pre>
+</p>
+<p>Similarly, since Jalview 2.5, group associated annotation can be defined by preceding the row definitions with the line:
+<pre>GROUP_REF	<em>group_name</em></pre>
+Group association is turned off for subsequent annotation rows by:
+<pre>GROUP_REF	<em>ALIGNMENT</em></pre>
+</p>
+<hr/>
+<p><strong><a name="refsandviews">VIEW_SETREF, VIEW_HIDECOL and HIDE_INSERTIONS</a></strong><br/>
+Since Jalview 2.8.3, the Annotations file has also supported the definition of views on the alignment, and definition of hidden regions.</p>
+<!-- <p>
+ <em>VIEW_DEF</em> allows the current view to be named according to the
+ first argument after the tab character. If a second argument is
+ provided, then a new view is created with the given name, and
+ properties.
+ </p> -->
+ <p>
+ <em>VIEW_SETREF</em> takes either a single sequence ID string, or a
+ numeric index (second argument), and attempts to assign a
+ corresponding sequence as the <a href="../features/refsequence.html">reference
+ sequence</a> for the alignment.
+ </p>
+ <em>VIEW_HIDECOLS</em> takes either a single argument consisting of a
+ comma separated series of integer pairs like
+ <em>3-4</em>. These integer pairs define columns (starting from the
+ left-hand column 0) that should be marked as hidden in the alignment
+ view.
+ </p>
+ <p>
+ <em>HIDE_INSERTIONS</em> takes a either a single sequence ID or a
+ numeric index, or no arguments. This command marks all gapped
+ positions in a specified sequence (either the one located by the
+ arguments, the current SEQUENCE_REF, or the reference sequence for the
+ view).
+ <hr/>
+<p><strong><a name="compatibility">COMPATIBILITY NOTES</a></strong><br/>
+ The interpretation of the COMBINE statement in <em>Version 2.8.1</em> was refined
+ so that only annotation line graphs with the given names ands the same
+ <strong>SEQUENCE_REF</strong> and <strong>GROUP_REF</strong> scope are grouped.</p>
+ <hr/>
+
+<p><strong><a name="exampleann">EXAMPLES</a></strong><br/>
+An example Annotation file is given below. Copy and paste the contents into a text file and load it onto the Jalview example protein alignment.</p>