diff --git a/.spelling b/.spelling
new file mode 100644
index 0000000000000000000000000000000000000000..0b5aca64943dfa6b724918aa34d2ad73cd47cfeb
--- /dev/null
+++ b/.spelling
@@ -0,0 +1,18 @@
+Biopython
+CIF
+FastA
+Jupyter
+MSA
+ModelCIF
+PAE
+PDB
+PPI
+coevolution
+modeled
+modeling
+multimer
+polypeptide
+pre
+repo
+reproducibility
+subdirectory
diff --git a/convert_to_modelcif.py b/convert_to_modelcif.py
index 0c7a11a4297673371c44fae3a3d76ca2e717e181..dbfae0fb71a9f9dfaaf8f180b73d0043b9fa775a 100755
--- a/convert_to_modelcif.py
+++ b/convert_to_modelcif.py
@@ -77,7 +77,7 @@ class _GlobalPTM(modelcif.qa_metric.Global, modelcif.qa_metric.PTM):
class _GlobalIPTM(modelcif.qa_metric.Global, modelcif.qa_metric.IpTM):
- # python-modelcif reads the first line of class-doc has description of a
+ # `python-modelcif` reads the first line of class-doc has description of a
# score, so we need to allow long lines, here.
# pylint: disable=line-too-long
"""Predicted protein-protein interface score, based on the TM-score score in [0,1]."""
@@ -111,7 +111,7 @@ class _LocalPairwisePAE(modelcif.qa_metric.LocalPairwise, _PAE):
class _Biopython2ModelCIF(modelcif.model.AbInitioModel):
- """Map Biopython PDB.Structure object to ihm.model"""
+ """Map Biopython `PDB.Structure()` object to `ihm.model()`."""
def __init__(self, *args, **kwargs):
"""Initialise a model"""
@@ -152,7 +152,7 @@ class _Biopython2ModelCIF(modelcif.model.AbInitioModel):
)
# local scores
- # iterate polypetide chains
+ # iterate polypeptide chains
# local PLDDT
i = 0
lpae = []
@@ -239,7 +239,7 @@ def _get_step_output_method_type(method_type, protocol_steps):
"""Get the output of a protocol step of a certain type."""
for step in protocol_steps:
if step.method_type == method_type:
- # modelcif.data.DataGroup is some kind of list
+ # `modelcif.data.DataGroup()` is some kind of list
if isinstance(step.output_data, list):
return step.output_data
return modelcif.data.DataGroup(step.output_data)
@@ -309,12 +309,16 @@ def _get_modelcif_protocol(
# add software w/o individual parameters
sw_grp.append(sw_dict[pss])
# add software with individual parameters
+ # Commented out code does not need spelling check, so disable
+ # it in Pylint
+ # pylint: disable=wrong-spelling-in-comment
# sw_grp.append(
# modelcif.SoftwareWithParameters(
# sw_dict[pss],
# [modelcif.SoftwareParameter(<name>, <value>)],
# )
# )
+ # pylint: enable=wrong-spelling-in-comment
# ToDo: make sure AlphaPulldown is first in the SoftwareGroup() list,
# AlphaFold second; that influences citation order in the ModelCIF
# file.
@@ -334,7 +338,7 @@ def _get_modelcif_protocol(
def _get_modelcif_ref_dbs(meta_json):
- """Get sequence DBs used for monomer features."""
+ """Get sequence databases used for monomer features."""
# vendor formatting for DB names/ URLs, extend on KeyError
db_info = {
"uniref90": {
@@ -435,13 +439,13 @@ def _store_as_modelcif(
# pickle files quickly exceeds reasonable storage use
# - 'modeling' (run_multimer_jobs.py), are the four modes reflected by
# the JSON data/ does the JSON data look different for each mode?
- # - are the scores only calculated by alpha-analysis.sif or do they
+ # - are the scores only calculated by `alpha-analysis.sif` or do they
# come out of run_multimer_jobs.py? Does this go into its own step?
# - what about including the tabular summary?
# - model selection: only if just a certain model is translated to
# ModelCIF, or mix it with scoring step?
#
- # - cancer-PPI-domains has 'coevolutin MSA'
+ # - cancer-PPI-domains has 'coevolution MSA'
# - then modelling
# - do we have an example with a split MSA & modelling step?
# - model selection like for Tara
@@ -457,7 +461,7 @@ def _store_as_modelcif(
)
)
- # write modelcif.System to file
+ # write `modelcif.System()` to file
# NOTE: this will dump PAE on path provided in add_scores
# -> hence we cheat by changing path and back while being exception-safe...
oldpwd = os.getcwd()
@@ -484,7 +488,7 @@ def _store_as_modelcif(
def _compress_cif_file(cif_file):
- """Compress cif file and delete original."""
+ """Compress CIF file and delete original."""
with open(cif_file, "rb") as f_in:
with gzip.open(cif_file + ".gz", "wb") as f_out:
shutil.copyfileobj(f_in, f_out)
@@ -495,7 +499,7 @@ def _get_model_details(cmplx_name: str, data_json: dict) -> str:
"""Get the model description."""
ap_versions = []
af2_version = None
- for mnmr in data_json["__meta__"]: # mnmr = monomer
+ for mnmr in data_json["__meta__"]: # `mnmr = monomer`
if (
data_json["__meta__"][mnmr]["software"]["AlphaPulldown"]["version"]
not in ap_versions
@@ -632,7 +636,7 @@ def _get_scores(cif_json: dict, scr_file: str) -> None:
def _get_software_data(meta_json: dict) -> list:
- """Turn meta data about software into modelcif.Software objects."""
+ """Turn meta data about software into `modelcif.Software()` objects."""
cite_hhsuite = ihm.Citation(
pmid="31521110",
title="HH-suite3 for fast remote homology detection and deep "
@@ -651,7 +655,7 @@ def _get_software_data(meta_json: dict) -> list:
],
doi="10.1186/s12859-019-3019-7",
)
- # {key from json: dict needed to produce sw entry plus internal key}
+ # {key from JSON: dict needed to produce software entry plus internal key}
sw_data = {
"AlphaFold": modelcif.Software(
"AlphaFold-Multimer",
@@ -811,7 +815,7 @@ def _get_protocol_steps(modelcif_json):
# model selection step <- ask if there is automated selection, if only
# manual, skip this step here?
- # ToDo: Example 1 in the GitHub repo has a 3rd step: "Evalutaion and
+ # ToDo: Example 1 in the GitHub repo has a 3rd step: "Evaluation and
# visualisation"
return protocol
@@ -980,6 +984,3 @@ if __name__ == "__main__":
# From former discussions:
# - including Jupyter notebooks would require adding the pickle files to the
# associated files (too much storage needed for that)
-
-# LocalWords: ToDo AlphaPulldown PAEs dir struct coevolution MSA py modeling
-# LocalWords: multimer sif Jupyter aa MSAs
diff --git a/pyproject.toml b/pyproject.toml
index f9bbc980b0989ac6396bd77332b9b7c90f98df8a..39913b661df304953bcf89905003eda628206c19 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -21,4 +21,19 @@ max-line-length = 81
[tool.pylint.deprecated_builtins]
# We want to use proper logging, so we can control *ALL* output bei the Abseil
# logger, hence: deprecate 'print'
-bad-functions = ["map", "filter", "print"]
\ No newline at end of file
+bad-functions = ["map", "filter", "print"]
+
+# Run the spell check every once in a while, having it enabled always, is too
+# annoying.
+#[tool.pylint.spelling]
+#max-spelling-suggestions = 4
+#
+#spelling-dict = "en_GB"
+#
+#spelling-ignore-comment-directives = "fmt: on,fmt: off,noqa:,noqa,nosec,isort:skip,mypy:"
+#
+#spelling-ignore-words = ""
+#
+#spelling-private-dict-file = ".spelling"
+#
+#spelling-store-unknown-words = false
\ No newline at end of file