This application is intended to evaluate coevolutionary structural predictions of fold-switching proteins. State-of-the art algorithms predict that these fold-switching proteins assume only one stable structure. We hypothesize that coevolutionary signatures are being missed. Fold-switching proteins have the ability to transition between two sets of stable secondary and tertiary structure. The approach successfully revealed coevolution of amino acid pairs uniquely corresponding to both conformations of 56 fold-switching proteins from distinct families.

References:

• Joseph W. Schafer and Lauren L. Porter,
  Evolutionary selection of proteins with two folds
  bioRxiv preprint doi: https://doi.org/10.1101/2023.01.18.524637

• dual_fold_coevolution Github page

• Module Name: DFC (see the modules page for more information)
• Unusual environment variables set
  • DFC_HOME  installation directory
  • DFC_BIN       executable directory

Allocate an interactive session and run the program. Sample session:

[user@biowulf]$ sinteractive 
[user@cig 3335 ~]$ module load DFC
[+] Loading singularity  4.0.1   on cn3335
[+] Loading DFC  20240124
[user@cn3335 ~]$ ACE.py -h 
...
usage: ACE.py [-h] [--msa MSA] [--Sub_Family_Parser SUB_FAMILY_PARSER] [--fasta FASTA] [--b_factor B_FACTOR] [--pdb1 PDB1]
              [--pdb2 PDB2] [--Extra EXTRA] [--Depth_max DEPTH_MAX] [--Depth_min DEPTH_MIN] [--Manual_PDB MANUAL_PDB]
              [--Manual_Pred MANUAL_PRED] [--Beg BEG] [--End END] [--Prediction_Offset PREDICTION_OFFSET]

optional arguments:
  -h, --help            show this help message and exit
  --msa MSA             MSA file in stockholm format
  --Sub_Family_Parser SUB_FAMILY_PARSER
                        Choose between parsing sub-family alignments by QID or E-value (E-value is from HMMER)
  --fasta FASTA         MSA file in fasta format
  --b_factor B_FACTOR   (y/n) ESMfold and Alphafold have pLDDT scores in the b_factor column. This flag colors the
                        dualfold contact map based on pLDDT, poor predictions are red and ok predictions are yellow.
  --pdb1 PDB1           PDB structure for the first conformation (remove hydrogens, ligands, and HOH
  --pdb2 PDB2           PDB structure for the second conformation (remove hydrogens, ligands, and HOH
  --Extra EXTRA         Output additional information on superposition, base case, and smallest subfamily alignment
  --Depth_max DEPTH_MAX
                        Control how many predictions are extracted from subfamily alignments (larger number = more
                        predictions)
  --Depth_min DEPTH_MIN
                        Control how many predictions are extracted from subfamily alignments (larger number = more
                        predictions)
  --Manual_PDB MANUAL_PDB
                        Turn off auto-align algorithm and manually align PDBs
  --Manual_Pred MANUAL_PRED
                        Turn off auto-align algorithm and manually align Predictions
  --Beg BEG             Remove x number of residues from the beginning of the msa
  --End END             Remove x number of residues from the end of the msa
  --Prediction_Offset PREDICTION_OFFSET
                        Used if a large number of residues are missing from the CTD (ex. 3ejh/3m7p alignment set to 225
[user@cn3335 ~]$ exit
user@biowulf]$