Add a library of all CCD residues separate from modified amino/nucleic acids (ParmEd#1315)

* Add templates for all CCD residues other than currently stored amino/nucleic acids

* Fix file name

* Fix issues in RDKit interconversions

* Fix copying of qualitative bond type

* Update dependencies.

* Do not arbitrarily override residue number

* Fix tests to work with newer OpenMM versions.

* Fix for newer scipy

Author: swails

.github/workflows/Test.yaml

@@ -13,11 +13,11 @@ jobs:
     strategy:
       matrix:
         os: ["ubuntu-latest"]
-        python-version: ["3.8", "3.9"]
+        python-version: ["3.9", "3.10"]
 
     steps:
     - name: Check out source code
-      uses: actions/checkout@v2
+      uses: actions/checkout@v3
       with:
         fetch-depth: 0
 
@@ -29,7 +29,7 @@ jobs:
         environment-file: devtools/environment-dev.yaml
         activate-environment: parmed-dev
         channels: conda-forge,bioconda
-        # mamba-version: "*"
+        mamba-version: "*"
 
     - name: Environment Information
       shell: bash -l {0}
@@ -38,7 +38,7 @@ jobs:
         conda list
 
     - name: Install and test
-      shell: bash -lex {0}
+      shell: bash -l {0}
       run: |
         bash -ex devtools/ci/install.sh
 

ccd.js

@@ -7,20 +7,39 @@ node ccd.js components.cif > components.json
 
 const { CIF } = require('molstar/lib/commonjs/mol-io/reader/cif');
 const fs = require('fs');
+const JSONStream = require('JSONStream');
 
-function isValidComponent(block) {
+function isLinkingMonomer(block) {
     const { chem_comp } = block.categories;
     const type = chem_comp.getField('type')?.str(0).toLowerCase() ?? '';
 
-    return type.endsWith('linking')
-        && (type.includes('dna') || type.includes('rna') || type.includes('peptide'));
+    return type.endsWith('linking') && type.includes('peptide');
+}
+
+function isNotLinkingMonomer(block) {
+    return !isLinkingMonomer(block);
 }
 
 function formatComponent(block) {
     const { chem_comp, chem_comp_atom, chem_comp_bond } = block.categories;
 
     // Atoms info
     const atoms = [];
+    const bonds = [];
+    let atomCount = chem_comp_atom?.rowCount ?? 0;
+    if (atomCount === 0) {
+        console.info("No atoms found for", chem_comp.getField('id')?.str(0));
+        return {
+            name: chem_comp.getField('id')?.str(0),
+            one_letter_code: chem_comp.getField('one_letter_code')?.str(0),
+            full_name: chem_comp.getField('name')?.str(0),
+            synonyms: chem_comp.getField('pdbx_synonyms')?.str(0),
+            type: chem_comp.getField('type')?.str(0),
+            formal_charge: chem_comp.getField('pdbx_formal_charge')?.int(0),
+            atoms,
+            bonds,
+        }
+    }
     for (let i = 0; i < chem_comp_atom.rowCount; i++) {
         atoms.push({
             name: chem_comp_atom.getField('atom_id').str(i),
@@ -33,9 +52,8 @@ function formatComponent(block) {
     }
 
     // Bonds info
-    const bonds = [];
     if (chem_comp_bond) {
-        for (let i = 0; i < chem_comp_bond.rowCount; i++) {
+        for (let i = 0; i < chem_comp_bond?.rowCount ?? 0; i++) {
             let bond = {
                 name_a: chem_comp_bond.getField('atom_id_1').str(i),
                 name_b: chem_comp_bond.getField('atom_id_2').str(i),
@@ -51,6 +69,7 @@ function formatComponent(block) {
         name: chem_comp.getField('id')?.str(0),
         one_letter_code: chem_comp.getField('one_letter_code')?.str(0),
         full_name: chem_comp.getField('name')?.str(0),
+        synonyms: chem_comp.getField('pdbx_synonyms')?.str(0),
         type: chem_comp.getField('type')?.str(0),
         formal_charge: chem_comp.getField('pdbx_formal_charge')?.int(0),
         atoms,
@@ -67,10 +86,21 @@ async function run() {
     }
 
     const components = parsed.result.blocks
-        .filter(isValidComponent)
+        .filter(isLinkingMonomer)
         .map(formatComponent);
 
-    console.log(JSON.stringify(components, null, 2));
+    var transformStream = JSONStream.stringify();
+    var outputStream = fs.createWriteStream('processed-components.json');
+    transformStream.pipe(outputStream);
+    components.forEach(transformStream.write);
+    transformStream.end();
+
+    outputStream.on(
+        "finish",
+        function handleFinish() {
+            console.log("Done");
+        }
+    );
 }
 
-run();
+run();

devtools/environment-dev.yaml

@@ -4,17 +4,14 @@ channels:
   - https://conda.anaconda.org/bioconda
 dependencies:
   - pandas
-  - nose
-  - openmm >=7.6.0,<7.7.0a
+  - openmm >=7.6.0
   - coverage
-  - nose-timer
   - netCDF4
-  - rdkit ==2020.09.4
+  - rdkit >=2020.09.4
   - nglview
-  - ambertools ==20.15
+  - ambertools >=20.15
   - networkx
   - lxml
-  - nose-timer
   - gromacs
   - pytest
   - pytest-cov

get_hybridization_from_rdkit.py

@@ -3,13 +3,14 @@
 extract the hybridization for each templated residue
 """
 import json
-from pathlib import Path
 from rdkit import Chem
+from tqdm import tqdm
 from pathlib import Path
 from parmed.rdkit import RDKit
-from parmed.modeller.standardtemplates import get_standard_residue_template_library
+from parmed.modeller.standardtemplates import get_standard_residue_template_library, get_nonstandard_ccd_residues
 
-residues = get_standard_residue_template_library()
+# residues = get_standard_residue_template_library()
+residues = get_nonstandard_ccd_residues()
 
 rdkit_map = dict()
 
@@ -18,16 +19,16 @@
     Chem.SanitizeMol(mol, Chem.SANITIZE_SETHYBRIDIZATION)
     rdkit_map[name] = mol
 
-with Path("ccd_residue_templates.json").open("r") as f:
+with Path("nonstandard_ccd_residue_templates.json").open("r") as f:
     data = json.load(f)
 
-for res in data:
+for res in tqdm(data):
     if res["name"] not in rdkit_map:
         print(f"Residue {res['name']} is not in the RDKit map")
         continue
     assert len(res["atoms"]) == rdkit_map[res["name"]].GetNumAtoms()
     for atom_data, atom in zip(res["atoms"], rdkit_map[res["name"]].GetAtoms()):
         atom_data["hybridization"] = atom.GetHybridization()
 
-with Path("ccd_residue_templates_with_hybridization.json").open("w") as f:
+with Path("nonstandard_ccd_residue_templates_with_hybridization.json").open("w") as f:
     json.dump(data, f)