Tool update: AlphaFold 2
The AlphaFold 2 tool has been updated in line with the latest best practice recommendations.
The AlphaFold 2 tool on Galaxy Australia has had an upgrade to ensure it is in line with best practice recommendations and provides all the reports and data visualisations required to publish your research.
Researchers from the Australian Structural Biology Computing community recently published a guide for presenting and sharing AlphaFold models in academic publications. Galaxy Australia’s Dr Cameron Hyde quickly set about upgrading the AlphaFold 2 tool, adding two key features:
- MSA sequence coverage plots (pictured below)
- Inclusion of PTM and iPTM scores in the model confidence scores output
During the upgrade, Cam also wrote a fully annotated Python script to resolve a commonly encountered error. Sometimes, the Protein Data Bank (PDB) mmCIF database is out of date with the other reference databases used by AlphaFold. The custom script cross-references your AlphaFold 2 databases and attempts to patch missing any .cif files from the RCSB database.
If you haven’t used AlphaFold 2 on Galaxy Australia yet, be sure to request access. It is freely available to all Australian researchers with an institutional email (per our reference list).
Plot showing sequence coverage in the multiple sequence alignment (MSA) against per-residue model confidence. These plots help the user decide whether regions of low model confidence are due to stereochemical properties of the protein, or just poor sequence coverage.