Merge pull request #655 from xchem/m2ms-1499-shortcode

Updated both short- and longcode (issue 1499)
xchem · Sep 6, 2024 · f6d6b6d · f6d6b6d
2 parents 852db67 + 7f31bb5
commit f6d6b6d
Show file tree

Hide file tree

Showing 2 changed files with 43 additions and 11 deletions.
diff --git a/viewer/download_structures.py b/viewer/download_structures.py
@@ -777,6 +777,25 @@ def _create_structures_dict(site_obvs, protein_params, other_params):
 
     # Read through zip_params to compile the parameters
     zip_contents: Dict[str, Any] = copy.deepcopy(zip_template)
+    site_obvs = site_obvs.annotate(
+        # would there be any point in
+        # a) adding a method to SiteObservation model_attr
+        # b) adding the value to database directly?
+        longlongcode=Concat(
+            F('experiment__code'),
+            Value('_'),
+            F('chain_id'),
+            Value('_'),
+            F('seq_id'),
+            Value('_'),
+            F('version'),
+            Value('_'),
+            F('canon_site_conf__canon_site__name'),
+            Value('+'),
+            F('canon_site_conf__canon_site__version'),
+            output_field=CharField(),
+        ),
+    )
     for so in site_obvs:
         for param in protein_params:
             if protein_params[param] is True:
@@ -799,9 +818,12 @@ def _create_structures_dict(site_obvs, protein_params, other_params):
                     for f in model_attr:
                         # here the model_attr is already stringified
                         try:
-                            exp_path = re.search(r"x\d*", so.code).group(0)  # type: ignore[union-attr]
-                        except AttributeError:
-                            logger.error('Unexpected shortcodeformat: %s', so.code)
+                            exp_path = so.experiment.code.split('-x')[1]
+                        except IndexError:
+                            logger.error(
+                                'Unexpected experiment code format: %s',
+                                so.experiment.code,
+                            )
                             exp_path = so.code
 
                         apath = Path('crystallographic_files').joinpath(exp_path)
@@ -843,7 +865,7 @@ def _create_structures_dict(site_obvs, protein_params, other_params):
                             apath.joinpath(
                                 Path(model_attr.name)
                                 .parts[-1]
-                                .replace(so.longcode, so.code)
+                                .replace(so.longlongcode, so.code)
                             )
                         )
                     else:
@@ -875,7 +897,9 @@ def _create_structures_dict(site_obvs, protein_params, other_params):
                             f'{ccp_path.stem}_crystallographic{ccp_path.suffix}'
                         )
                         archive_path = str(
-                            apath.joinpath(path.parts[-1].replace(so.longcode, so.code))
+                            apath.joinpath(
+                                path.parts[-1].replace(so.longlongcode, so.code)
+                            )
                         )
 
                         afile = [

diff --git a/viewer/target_loader.py b/viewer/target_loader.py
@@ -1106,6 +1106,8 @@ def process_canon_site(
         conf_sites_ids = extract(key="conformer_site_ids", return_type=list)
         ref_conf_site_id = extract(key="reference_conformer_site_id")
 
+        centroid_res = f"{centroid_res}_v{version}"
+
         fields = {
             "name": canon_site_id,
             "version": version,
@@ -1311,7 +1313,8 @@ def process_site_observation(
         experiment = experiments[experiment_id].instance
 
         longcode = (
-            f"{experiment.code}_{chain}_{str(ligand)}_{str(version)}_{str(v_idx)}"
+            # f"{experiment.code}_{chain}_{str(ligand)}_{str(version)}_{str(v_idx)}"
+            f"{experiment.code}_{chain}_{str(ligand)}_v{str(version)}"
         )
         key = f"{experiment.code}/{chain}/{str(ligand)}"
         v_key = f"{experiment.code}/{chain}/{str(ligand)}/{version}"
@@ -1758,7 +1761,8 @@ def process_bundle(self):
                         ]
                         # iter_pos = next(suffix)
                         # code = f"{code_prefix}{so.experiment.code.split('-')[1]}{iter_pos}"
-                        code = f"{code_prefix}{so.experiment.code.split('-')[1]}{next(suffix)}"
+                        # code = f"{code_prefix}{so.experiment.code.split('-')[1]}{next(suffix)}"
+                        code = f"{code_prefix}{so.experiment.code.split('-x')[1]}{next(suffix)}"
 
                         # test uniqueness for target
                         # TODO: this should ideally be solved by db engine, before
@@ -1858,7 +1862,10 @@ def process_bundle(self):
 
         numerators = {}
         cat_conf = TagCategory.objects.get(category="ConformerSites")
-        for val in canon_site_conf_objects.values():  # pylint: disable=no-member
+        for val in canon_site_conf_objects.values():  # pylint:
+            # disable=no-member problem introduced with the sorting of
+            # canon sites (issue 1498). objects somehow go out of sync
+            val.instance.refresh_from_db()
             if val.instance.canon_site.canon_site_num not in numerators.keys():
                 numerators[val.instance.canon_site.canon_site_num] = alphanumerator()
             prefix = (
@@ -2163,7 +2170,8 @@ def _tag_observations(
             so_group.save()
 
         name = f"{prefix} - {tag}" if prefix else tag
-        short_tag = name if short_tag is None else f"{prefix} - {short_tag}"
+        tag = tag if short_tag is None else short_tag
+        short_name = name if short_tag is None else f"{prefix} - {short_tag}"
 
         try:
             so_tag = SiteObservationTag.objects.get(
@@ -2175,14 +2183,14 @@ def _tag_observations(
             so_tag.mol_group = so_group
         except SiteObservationTag.DoesNotExist:
             so_tag = SiteObservationTag(
-                tag=short_tag,
+                tag=tag,
                 tag_prefix=prefix,
                 upload_name=name,
                 category=category,
                 target=self.target,
                 mol_group=so_group,
                 hidden=hidden,
-                short_tag=short_tag,
+                short_tag=short_name,
             )
 
         so_tag.save()