CIF improvement

README.md

@@ -95,7 +95,7 @@ cd -
 export PYO3_CROSS_INCLUDE_DIR="HACK"
 export PYO3_CROSS_LIB_DIR="HACK"
 rm -rf .pyodide-xbuildenv
-pyodide build --build-dependencies -o dist_pyodide
+pyodide build -o dist_pyodide
 cp dist_pyodide/*.whl test_wasm/wheels/
 cd test_wasm
 python3 -m http.server

moleculekit/molecule.py

@@ -21,6 +21,68 @@ def __str__(self):
         return repr(self.value)
 
 
+_originalResname = {
+    "ARG": "ARG",
+    "AR0": "ARG",
+    "HIS": "HIS",
+    "HID": "HIS",
+    "HIE": "HIS",
+    "HIP": "HIS",
+    "HSD": "HIS",
+    "HSE": "HIS",
+    "HSP": "HIS",
+    "LYS": "LYS",
+    "LSN": "LYS",
+    "LYN": "LYS",
+    "ASP": "ASP",
+    "ASH": "ASP",
+    "GLU": "GLU",
+    "GLH": "GLU",
+    "SER": "SER",
+    "THR": "THR",
+    "ASN": "ASN",
+    "GLN": "GLN",
+    "CYS": "CYS",
+    "CYM": "CYS",
+    "CYX": "CYS",
+    "SEC": "SEC",
+    "GLY": "GLY",
+    "PRO": "PRO",
+    "ALA": "ALA",
+    "VAL": "VAL",
+    "ILE": "ILE",
+    "LEU": "LEU",
+    "MET": "MET",
+    "PHE": "PHE",
+    "TYR": "TYR",
+    "TRP": "TRP",
+    "G": "G",
+    "G5": "G",
+    "G3": "G",
+    "C": "C",
+    "C5": "C",
+    "C3": "C",
+    "U": "U",
+    "U5": "U",
+    "U3": "U",
+    "A": "A",
+    "A5": "A",
+    "A3": "A",
+    "T": "T",
+    "DG": "G",
+    "DG5": "G",
+    "DG3": "G",
+    "DC": "C",
+    "DC5": "C",
+    "DC3": "C",
+    "DA": "A",
+    "DA5": "A",
+    "DA3": "A",
+    "DT": "T",
+    "DT5": "T",
+    "DT3": "T",
+}
+
 _residueNameTable = {
     "ARG": "R",
     "AR0": "R",
@@ -2238,6 +2300,25 @@ def getNeighbors(self, idx):
     def _ix(self, resid, name):
         return np.where((self.name == name) & (self.resid == resid))[0][0]
 
+    def toOpenFFMolecule(self):
+        import tempfile
+        from moleculekit.smallmol.smallmol import SmallMol
+        from openff.toolkit.topology import Molecule as OFFMolecule
+
+        with tempfile.TemporaryDirectory() as tmpdir:
+            molc = self.copy()
+            # for i in range(molc.numAtoms):
+            #     molc.name[i] = molc.name[i] + str(i)
+            pdbfile = os.path.join(tmpdir, "mol.pdb")
+            smiles = SmallMol(molc, fixHs=False).toSMILES(
+                explicitHs=True, kekulizeSmile=True
+            )
+            molc.write(pdbfile)
+            offmol = OFFMolecule.from_pdb_and_smiles(pdbfile, smiles=smiles)
+            assert np.array_equal(molc.name, [x.name for x in offmol.atoms])
+
+            return offmol
+
 
 class UniqueAtomID:
     _fields = ("name", "altloc", "resname", "chain", "resid", "insertion", "segid")

moleculekit/readers.py

@@ -1933,7 +1933,7 @@ def CIFread(filename, frame=None, topoloc=None, zerowarning=True):
     dataObj = myDataList[0]
 
     def fixDefault(val, dtype):
-        if val == "?":
+        if val in ("?", "."):
             if dtype == float or dtype == int:
                 val = 0
             if dtype == str:
@@ -2336,7 +2336,31 @@ def sdf_generator(sdffile):
                 sdfstr = ""
 
 
+# This is a hack to fix the URL fetching of MMTF in pyodide
+def get_raw_data_from_url(pdb_id, reduced=False):
+    """ " Get the msgpack unpacked data given a PDB id.
+
+    :param pdb_id: the input PDB id
+    :return the unpacked data (a dict)"""
+    from mmtf.api.default_api import get_url, ungzip_data, _unpack, urllib2
+
+    url = get_url(pdb_id, reduced)
+    request = urllib2.Request(url)
+    request.add_header("Accept-encoding", "gzip")
+    response = urllib2.urlopen(request)
+    if response.info().get("Content-Encoding") == "gzip":
+        data = ungzip_data(response.read())
+    else:
+        # Fixed here from original data = response.read()
+        data = response
+    return _unpack(data)
+
+
 def MMTFread(filename, frame=None, topoloc=None, validateElements=True):
+    from mmtf.api import default_api
+
+    # Monkey-patch the function to fix bug in data = response.read() which should not have read()
+    default_api.get_raw_data_from_url = get_raw_data_from_url
     from mmtf import fetch, parse_gzip, parse
 
     if len(filename) == 4 and not os.path.isfile(filename):