Proposed fix for issue #47

[alex-selvita]

Added support for residues from extended IUPAC set and UNK (unknown) residues encodings from 3 letter to 1 letter sequence.

Fixes #47

[jvkersch]

@alkorolyov-selvita Thanks for the fix! This does the job but has the side-effect of modifying the IUPACData dictionary, which "belongs to" BioPython (so that if this dictionary is used elsewhere in a user's code, the modification will also be visible). I would propose having an explicit fix: e.g. check for residue.resname == 'UNK' on the line below, and then insert X or the appropriate 1-letter code in the sequence:

tmtools/tmtools/io.py

Line 119 in 10dde1d

seq.append(protein_letters_3to1[residue.resname])

.

[jvkersch]

@alkorolyov-selvita Thanks for the fix. I've left one comment; could you also add a test to exercise the fix?

[alex-selvita]

Hello @jvkersch thanks for your comments, sure I will update the code and try to add some basic tests for new changes.

[jvkersch]

@alkorolyov-selvita Did you have time to look into this? I think there are two issues, which are easy to address.

The first is that (apparently) PDB uses uppercase for the residues, whereas the keys in protein_letters_3to1_extended are lowercase.

>>> from Bio.PDB.Polypeptide import protein_letters_3to1
>>> protein_letters_3to1
{'ALA': 'A', 'CYS': 'C', 'ASP': 'D', 'GLU': 'E', 'PHE': 'F', 'GLY': 'G', 'HIS': 'H', 'ILE': 'I', 'LYS': 'K', 'LEU': 'L', 'MET': 'M', 'ASN': 'N', 'PRO': 'P', 'GLN': 'Q', 'ARG': 'R', 'SER': 'S', 'THR': 'T', 'VAL': 'V', 'TRP': 'W', 'TYR': 'Y'}
>>> from Bio.Data.IUPACData import protein_letters_3to1_extended
>>> protein_letters_3to1_extended
{'Ala': 'A', 'Cys': 'C', 'Asp': 'D', 'Glu': 'E', 'Phe': 'F', 'Gly': 'G', 'His': 'H', 'Ile': 'I', 'Lys': 'K', 'Leu': 'L', 'Met': 'M', 'Asn': 'N', 'Pro': 'P', 'Gln': 'Q', 'Arg': 'R', 'Ser': 'S', 'Thr': 'T', 'Val': 'V', 'Trp': 'W', 'Tyr': 'Y', 'Asx': 'B', 'Xaa': 'X', 'Glx': 'Z', 'Xle': 'J', 'Sec': 'U', 'Pyl': 'O'}

The second is that the protein_letters_3to1_extended dictionary should not be modified by this code. There are a few workarounds: one is to create a copy of protein_letters_3to1_extended internal to this library and modify that. That would also allow you to do something about the capitalization:

_PROTEIN_LETTERS_3TO1_INTERNAL = {k.upper(): v for (k, v) in protein_letters_3to1.items()}
_PROTEIN_LETTERS_3TO1_INTERNAL["UNK"] = "X"

[alex-selvita]

Hello @jvkersch ,

Sorry for long reply, I was struggling a little bit to properly configure the dev environment. For the moment I just reimplemented the get_resiude_data() locally to include the necesseray residue codes. I will try to get back to this issue alter.

[jvkersch]

@alkorolyov-selvita Let me know if you have any questions about setting up a dev environment. Personally, I use an empty Conda environment and then I install the package in editable mode (pip install -e . -v). My editor picks up the environment, and the only thing I have to do manually is recompile the C++ code when it changes (but this should not affect you).