1. Create an alignment fileΒΆ
1. Create a directory that will contain the results produced by Polyphony and move to it.
2. Run polyphony_create_fasta_file.py to download pdb files and create a sequence alignment file for a set of homologous or identical proteins. This executable should appear on your path on opening the first new window post installation.
polyphony_create_fasta_file.py -p 1PW6 -c A -s 95
The -p option specifies a pdb code of protein you are interested in, -c the chain letter and -s the percentage similarity cutoff -h for more help. This should a produce a file called clust_1PW6_A_95.fasta. Please be aware that if the program tries to download a lot of files then the ftp server complains (e.g. IOError: [Errno ftp error] 421 Too many connections (5) from this IP). If this happens you will need to rerun the script until all your files have been downloaded. The location of the ftp server is specified in polyphony.cfg.
3. Please view this alignment using your favourite multiple alignment software package e.g. clustalx. It might be that you need to do a sequence alignment or remove columns that are all gaps. If your structures are all of the same protein, then a structure alignment should be done instead. If this is the case, then fasta file produced by whatever should be checked using polyphony_create_fasta_file.py with the -f option. This is to ensure the provided sequences match the residues in the structures used (as interpreted by Polyphony/Biopython). All sequences should be have identifiers of the form pdbcode_chainletter e.g. 1PW6_A.