PyMissense generates the pathogenicity plot and modified pdb as shown in the AlphaMissense paper for custom proteins.

AlphaMissense allows you to identify regions in your amino acid chain that are critical for protein function. This script does two things:

1. It generates a plot similar to Figure 3D of the AlphaMissense paper

   

2. It generates a modified PDB file where the temperature factor (bfactor) is replaced by the pathogenicity predicted by AlphaMissense, allowing the effect to be visualised with Chimerax, as in Figure 3E of the paper:

   

pip install pymissense

General usage is:

usage: pymissense[-h] [--tsv TSV] [--pdbpath PDBPATH] [--maxacid MAXACID] uniprot_id output_path

AlphaMissense plot and pdb generator

positional arguments:
  uniprot_id         UNIPROT ID
  output_path        Output folder

options:
  -h, --help         show this help message and exit
  --tsv TSV          You can provide the path to the tsv file if you want to skip the download. (default: None)
  --pdbpath PDBPATH  If defined, it will write the pathogenicity as bfactor in that PDB. If its not defined or not existing it will instead download the alphafold predicted PDB (default: None)
  --maxacid MAXACID  Maximum squence number to use in the plot. (default: None)

You can give the optional argument --pdbpath if you want to use an experimental PDB, otherwise it will download the alphafold predicted PDB instead. With --tsv you can provide the decompressed AlphaMissense database, so that pymissense does not need to download it.

For example, to reproduce Figure 3D (the middle one) and generate the PDB shown in Figure 3E do:

wget https://files.rcsb.org/download/7UPI.pdb
pymissense Q9UQ13 out --maxacid 200 --pdbpath 7upi.pdb 

Note that only the first 200 amino acids are shown in the plots and the pathogenicity is shown with the experimental PDB 7upi.

This script was developed in collaboration with Tobias Raisch