diff --git a/.gitignore b/.gitignore
index c527082c72996ed22aff6b5a497c366799093e9c..3610828a887ca275f0672225a55e13a55742a8de 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,7 +1,12 @@
# Don't have Emac's backup files
*~
+# ignore test scripts
+biop-test.py
+junk.py
+
# ignore some files used for testing
1ake.1.pdb
3lre.3.A.pdb
6xne.pdb
+cmp.cif
diff --git a/convert_to_modelcif.py b/convert_to_modelcif.py
index 91a5176e512f6087bc734d5133be8817bd897e2a..fc30f1cd719ceb2f55ce0404867df9f025f1472e 100755
--- a/convert_to_modelcif.py
+++ b/convert_to_modelcif.py
@@ -20,6 +20,7 @@ from Bio.PDB.Structure import Structure as BioStructure
from absl import app, flags, logging
import numpy as np
+import ihm.citations
import modelcif
import modelcif.associated
import modelcif.dumper
@@ -120,8 +121,13 @@ class _Biopython2ModelCIF(modelcif.model.AbInitioModel):
occupancy=atm.occupancy,
)
- def add_scores(self, scores_json, entry_id, file_prefix):
+ def add_scores(self, scores_json, entry_id, file_prefix, sw_dct):
"""Add QA metrics"""
+ _GlobalPLDDT.software = sw_dct["alphafold"]
+ _GlobalPTM.software = sw_dct["alphafold"]
+ _GlobalIPTM.software = sw_dct["alphafold"]
+ _LocalPLDDT.software = sw_dct["alphafold"]
+ _LocalPairwisePAE.software = sw_dct["alphafold"]
# global scores
self.qa_metrics.extend(
(
@@ -254,9 +260,14 @@ def _store_as_modelcif(
name="ToDo: Model <N> (ranked #<M>)",
)
+ # create software list from feature metadata
+ sw_dct = _get_software_data(data_json["__meta__"])
+
# process scores
mdl_file = os.path.splitext(os.path.basename(mdl_file))[0]
- system.repositories.append(model.add_scores(data_json, system.id, mdl_file))
+ system.repositories.append(
+ model.add_scores(data_json, system.id, mdl_file, sw_dct)
+ )
system.model_groups.append(modelcif.model.ModelGroup([model]))
@@ -316,9 +327,16 @@ def _compress_cif_file(cif_file):
def _get_model_details(cmplx_name: str, data_json: dict) -> str:
"""Get the model description."""
ap_versions = []
- for mnmr in data_json["__meta__"]:
- if data_json["__meta__"][mnmr]["ap_version"] not in ap_versions:
- ap_versions.append(data_json["__meta__"][mnmr]["ap_version"])
+ for mnmr in data_json["__meta__"]: # mnmr = monomer
+ if (
+ data_json["__meta__"][mnmr]["software"]["alphapulldown"]["version"]
+ not in ap_versions
+ ):
+ ap_versions.append(
+ data_json["__meta__"][mnmr]["software"]["alphapulldown"][
+ "version"
+ ]
+ )
# ToDo: fetch AF2 version/ have it in metadata JSON
return (
@@ -348,7 +366,13 @@ def _get_feature_metadata(
# ToDo: make sure that its always ASCII
with open(feature_json, "r", encoding="ascii") as jfh:
jdata = json.load(jfh)
- modelcif_json["__meta__"][mnmr]["ap_version"] = jdata["version"]
+ modelcif_json["__meta__"][mnmr]["software"] = jdata["binaries"]
+ modelcif_json["__meta__"][mnmr]["software"]["alphapulldown"] = {
+ "version": jdata["version"]
+ }
+ modelcif_json["__meta__"][mnmr]["software"]["alphafold"] = {
+ "version": jdata["AlphaFold version"]
+ }
return cmplx_name
@@ -418,7 +442,6 @@ def _get_scores(cif_json: dict, scr_file: str) -> None:
"""Add scores to JSON data."""
with open(scr_file, "rb") as sfh:
scr_dict = pickle.load(sfh)
- # dict_keys(['distogram', 'experimentally_resolved', 'masked_msa', 'num_recycles', 'structure_module', 'aligned_confidence_probs', 'max_predicted_aligned_error', 'ranking_confidence'])
# Get pLDDT as a list, the global pLDDT is the average, calculated on the
# spot.
cif_json["plddt"] = scr_dict["plddt"]
@@ -427,6 +450,79 @@ def _get_scores(cif_json: dict, scr_file: str) -> None:
cif_json["pae"] = scr_dict["predicted_aligned_error"]
+def _get_software_data(meta_json: dict) -> list:
+ """Turn meta data about software into modelcif.Software objects."""
+ # {key from json: dict needed to produce sw entry plus internal key}
+ sw_data = {
+ "jackhmmer": None,
+ "hhblits": None,
+ "hhsearch": None,
+ "hmmsearch": None,
+ "hmmbuild": None,
+ "kalign": None,
+ "alphapulldown": None,
+ "alphafold": modelcif.Software(
+ "AlphaFold-Multimer",
+ "model building",
+ "Structure prediction",
+ "https://github.com/deepmind/alphafold",
+ "package",
+ None,
+ ihm.Citation(
+ pmid=None,
+ title="Protein complex prediction with AlphaFold-Multimer.",
+ journal="bioRxiv",
+ volume=None,
+ page_range=None,
+ year=2021,
+ authors=[
+ "Evans, R.",
+ "O'Neill, M.",
+ "Pritzel, A.",
+ "Antropova, N.",
+ "Senior, A.",
+ "Green, T.",
+ "Zidek, A.",
+ "Bates, R.",
+ "Blackwell, S.",
+ "Yim, J.",
+ "Ronneberger, O.",
+ "Bodenstein, S.",
+ "Zielinski, M.",
+ "Bridgland, A.",
+ "Potapenko, A.",
+ "Cowie, A.",
+ "Tunyasuvunakool, K.",
+ "Jain, R.",
+ "Clancy, E.",
+ "Kohli, P.",
+ "Jumper, J.",
+ "Hassabis, D.",
+ ],
+ doi="10.1101/2021.10.04.463034",
+ ),
+ ),
+ }
+ for data in meta_json.values():
+ for sftwr, version in data["software"].items():
+ if sftwr not in sw_data:
+ raise RuntimeError(
+ "Unknown software found in meta data: " + f"'{sftwr}'"
+ )
+ version = version["version"]
+ if sw_data[sftwr] is not None:
+ if sw_data[sftwr].version is not None:
+ if sw_data[sftwr].version != version:
+ raise RuntimeError(
+ "Software versions differ for "
+ + f"'{sftwr}': '{sw_data[sftwr].version}' vs. "
+ + f"'{version}'"
+ )
+ sw_data[sftwr].version = version
+
+ return sw_data
+
+
def alphapulldown_model_to_modelcif(
cmplx_name: str,
mdl_file: str,