"Mond" is the German equivalent to the meaning of the surname Ramachandran. The Ramachandran plot translates to moon plot. The dihedral angles ϕ (phi) vs. ψ (psi), which are free angles of rotation of an amino acid in a protein backbone, can be plotted in a Ramachandran plot. The peptide bond angle ω (omega) is restricted to mostly 180° due to the partial double-bond nature of the peptide connection.
Three regions in the Ramachandran are often populated: the top left with β-strands and sheets, the bottom left with left-handed α-helices, and the right middle with right-handed α-helices. The β-population is dominated by electrostatic forces, while the α-regions are preferred due to steric benefits.
Play around with your favourite canonical amino acid background! Then, choose non-canonical amino acid candidates and compare their populated regions across structures. Do they explore novel parts of the fold landscapes? Alternatively, focus on only one structure. See how the fold from N-terminus to C-terminus places the amino acids in backbone.
If you have a specific feature or functionality in mind that you would like to see implemented in this app, please let us know! We are always looking for ways to improve and enhance the user experience. Please send an email to alexandra.waldherr@tuebingen.mpg.de or gesa-laura.freimann@tuebingen.mpg.de.
Non-canonical amino acids are best filtered using three-letter codes. Here, all amino-acid like compounds from the wwPDB Chemical Component Dictionary (1913 ncAA and synthetic linkers, see the Chemical properties tab for sets of subclasses) are listed and enriched in which PDB-IDs, which specific chain, which organism, by which method and at which resolution the proteins include these backbone building block.
Look for your favorite non-canonical amino acid, look for your dataset of interest! We provide two sequences: The traditional one-letter sequence (e.g. as used by AlphaFold and .fasta-files) where all non-canonical compound are unspecifically hidden behind an X. For improvement, the column 'Unmasked Sequence' provides three-letter codes (XXX) bracketed instead of the one-letter X position. Your filtered dataset can be printed and downloaded under attribution to the authors. Please respect correct attributions. 😊 The dataset is licensed under the Creative Commons License 4.0. We advise attribution to the RSCB PDB mirror (https://pdb101.rcsb.org/more/how-to-cite) and the wwPDB Chemical Component Dictionary (https://www.wwpdb.org/data/ccd) out of love to the amazing work done before us.
| SMILES | ID | Name | Count | PDB ID | Description | Organism | Experiment | Resolution | Unmasked Sequence | Old Masked Sequence (X or parent) | Non-Canonicals in Sequence |
|---|
In the two inset windows, you can compare two chemical structures. A non-canonical amino acid to a canonical, or two non-canonicals, can be compared, as you wish. If you fetched an ncAA in the Ramachandran plot tab, the same ncAA will be displayed on the left. You can adjust which two compounds you would like to compare by entering SMILES into the free text fields below each window (OpenEye stereochemical SMILES, to obtain them use the table below).
In the table, SMILES and Tanimoto similarity for non-canonical amino acids are given. You can search for an ncAA by its ID. Each column lists the compounds Tanimoto similarities after MORGAN bits (see RDkit) to a canonical amino acid. The table can be sorted by high similarity (value: 1) or low similarity (value: 0). Click on the column header to sort. When downloading the table, take care. It is separated by ;.
Here, the protein structure as fetched in the first tab for the Ramachandran plot is displayed (default: 1UJS). Standard canonicals are displayed gray. When loading a structure with non-canonicals, the non-canonicals are highlighted red and fully displayed as sticks. The controls work analogously to the PDB. Note: If no structure appears, this is due to discontinuity in a chain. Check for ? in the Ramachandran plot tab chain list to see. :)