Allow for sites with insertion codes

CHANGELOG.md

@@ -165,4 +165,23 @@
 
 ### Removed
 
+- N/A
+
+
+## [1.2.0] - 2024-02-13
+
+### Added
+
+- N/A
+
+### Changed
+
+- It's now possible to include `protein_sites` proceed by insertion codes like (i.e. 52A, 214B, etc...). Previously, `protein_sites` could only be numeric and non-numeric sites were ignored by the `dms-viz`.
+
+### Deprecated
+
+- N/A
+
+### Removed
+
 - N/A

configure_dms_viz/configure_dms_viz.py

@@ -112,12 +112,13 @@ def format_sitemap_data(sitemap_df, mut_metric_df, included_chains):
     """Check that the sitemap data is in the correct format.
 
     This data should be a pandas.DataFrame with the following columns:
-    - reference_site: The site number in the reference sequence that
+    - reference_site: (numeric or string) The site number in the reference sequence that
         corresponds to the reference site in the mutation dataframe
-    - sequential_site: The order of the site in the protein sequence and
+    - sequential_site: (numeric) The order of the site in the protein sequence and
         on the x-axis of the visualization
-    - protein_site: [Optional] The site number in the protein structure if
-        different from the reference site
+    - protein_site: [Optional] (numeric or string) The site number in the protein structure if
+        different from the reference site. This can include insertion codes and
+        therefore can be a string.
 
     Parameters
     ----------
@@ -182,9 +183,7 @@ def is_numeric(value):
             message="\n'protein_site' column is not present in the sitemap. Assuming that the reference sites correspond to protein sites.\n",
             fg="yellow",
         )
-        sitemap_df["protein_site"] = sitemap_df["reference_site"].apply(
-            lambda y: y if is_numeric(y) else ""
-        )
+        sitemap_df["protein_site"] = sitemap_df["reference_site"].apply(lambda y: y)
         # Check how many of the protein sites are thrown out
         num_empty_protein_sites = (sitemap_df["protein_site"] == "").sum()
         if num_empty_protein_sites > 0.10 * len(sitemap_df):
@@ -193,22 +192,11 @@ def is_numeric(value):
                 fg="red",
             )
 
-    else:
-        # Make sure that the provided protein column has no invalid values
-        if (
-            not sitemap_df["protein_site"]
-            .apply(lambda y: y == "" or is_numeric(y))
-            .all()
-        ):
-            raise ValueError(
-                "The protein_site column of the sitemap contains invalid values that cannot be coerced into a numeric form."
-            )
-
     # If the sitemap doesn't already have a column for chains, add it
     if "chains" not in sitemap_df.columns:
         # Add the included chains to the sitemap dataframe if there are any
         sitemap_df["chains"] = sitemap_df["protein_site"].apply(
-            lambda y: included_chains if is_numeric(y) else ""
+            lambda y: included_chains
         )
 
     # Drop the columns that aren't needed for the visualization

configure_dms_viz/pdb_utils.py

@@ -173,10 +173,19 @@ def check_wildtype_residues(structure, mut_metric_df, sitemap_df, excluded_chain
         total_sites += 1
         matches_wildtype_at_site = []
         site_not_in_structure = []
+        # Convert the structure object a dictionary to include insertion codes in the residue id
+        structure_dict = {
+            chain: {
+                (str(residue.id[1]) + residue.id[2]).strip(): seq1(residue.resname)
+                for residue in structure[0][chain]
+                if residue.id[0] == " "
+            }
+            for chain in chains
+        }
         for chain in chains:
             try:
-                residue = structure[0][chain][int(site)]
-                if seq1(residue.resname) == wildtype.upper():
+                residue = structure_dict[chain][str(site)]
+                if residue == wildtype.upper():
                     matches_wildtype_at_site.append(True)
                 else:
                     matches_wildtype_at_site.append(False)

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "configure-dms-viz"
-version = "1.1.1"
+version = "1.2.0"
 description = "Configure your data for visualization with dms-viz.github.io"
 authors = ["Will Hannon <[email protected]>"]
 license = "MIT"