samberry19 / bioviper Goto Github PK

Hey Sam! It would be nice at some point to support reading structure coordinates in .cif format, which are becoming increasingly common. It seems that Bio.PDB does have a an MMCIFParser() object, so it'd just be a matter of interfacing with that (I think).

In the mean time, it would be good for the bioviper.readPDB() method to check for file extension and raise a helpful error when you try to use it on a different file format!

I'm happy to take a stab at both of these when I get a chance.

I just did the following:

conda create -n bvtest python=3.10
conda activate bvtest
pip install bioviper
conda list

and got the following:

# packages in environment at /Users/dennisbrookner/miniconda3/envs/bvtest:
#
# Name                    Version                   Build  Channel
bioviper                  0.2.0                    pypi_0    pypi
bzip2                     1.0.8                h0d85af4_4    conda-forge
ca-certificates           2023.7.22            h8857fd0_0    conda-forge
libffi                    3.4.2                h0d85af4_5    conda-forge
libsqlite                 3.43.0               h58db7d2_0    conda-forge
libzlib                   1.2.13               h8a1eda9_5    conda-forge
ncurses                   6.4                  hf0c8a7f_0    conda-forge
openssl                   3.1.3                h8a1eda9_0    conda-forge
pip                       23.2.1             pyhd8ed1ab_0    conda-forge
python                    3.10.12         had23ca6_0_cpython    conda-forge
readline                  8.2                  h9e318b2_1    conda-forge
setuptools                68.2.2             pyhd8ed1ab_0    conda-forge
tk                        8.6.12               h5dbffcc_0    conda-forge
tzdata                    2023c                h71feb2d_0    conda-forge
wheel                     0.41.2             pyhd8ed1ab_0    conda-forge
xz                        5.2.6                h775f41a_0    conda-forge

so it seems that the requirements listed in requirements.txt are not getting installed by pip!

Additionally, it seems to me that tqdm is a dependency but is not listed in requirements.txt.

I wrote the pdb module under the assumption that "disordered" residues do not have any corresponding structured atoms in the PDB file. However, I wasn't taking into account the fact that atoms can be disordered because they have multiple conformations.

For the time being, any "disordered atom" associated with a "disordered residue" (per Biopython's nomenclature) is removed from the .atoms attribute and from all downstream analyses, so that the .ordered_sequence attribute should exactly match the residues represented by, say, the .distance_matrix() function.

I need to think about this some more, however, because it may be that there are cases where we don't want this, or there might be some way of using the multiple conformations - in theory it would be nice not to just throw this information out. It's possible that I should add a flag for loading in a protein, and have a warning appear if multiple-conformation residues are thrown out, so that they can be reloaded at the coder's risk.