+ <p>Predicted Alignment Error (PAE) matrices are produced by
+ deep-learning based 3D-structure prediction pipelines such as
+ AlphaFold. They reflect how reliably two parts of a model have been
+ positioned in space, by giving for each residue the likely error (in
+ Ångstroms) between that residue and every other modelled
+ position the pair of residues' real relative position, if the model
+ and real 3D structure were superimposed at that residue.</p>
+ <p>
+ Jalview visualises PAE matrices as an alignment annotation track,
+ shaded from dark green to white, similar to the encoding used on the
+ EBI-AlphaFold website (see <a
+ href="https://alphafold.ebi.ac.uk/entry/O04090">O04090 3D model</a>
+ at EBI-AlphaFoldDB).
+ </p>
+ <div style="display:flex; flex-wrap:wrap;"align="center" width="100%">
+ <figure><img src="../structures/epas1_annotdetail.png" height="300"/><figcaption>Alignment of EPAS1 homologs from Human, Rat and Cow<br/>with predicted alignment error tracks</figcaption></figure>
+ <figure><img src="../structures/epas1_pae_ebiaf.png" height="300"/><figcaption>Predicted Alignment Error Matrix<br/>from <a href="https://alphafold.ebi.ac.uk/entry/Q99814">https://alphafold.ebi.ac.uk/entry/Q99814</a></figcaption></figure>
+ </div>
+ <p>
+ <strong>Importing PAE Matrices</strong>
+ </p>