Mol2 residue and molecule info for sites. (#671)

* modify mbuild converter to flatten compound hierarchy

* Add change method to determine molecule_group

* Add new label ("molecule" and "group") to help with the conversion

* fix typo and paritally update tests

* update from_mbuild tests

* Add molecule_number update docs
Also start translating residue info

* Make site.group optional string

* add missing var

* WIP: Remodel the labeling system, add parse_label
Combine residue name and number to be residue,
combine molecule name and number to be molecule.
Add parse label for from_mbuild method, but as of
right now, has really really bad performance (need
to rethink the logic here)

* remove the cloning step, improve performance

* add unit test

* update residue handling in convert parmed

* change method to parse label when convert from mbuild

* include missing import

* modify iter_site_by_residue, add iter_site_by_molecule

* update to_mbuild to match with new construct

* fix edge case when looking up None molecule and residue

* fix mol2 reader for new residue and molecule setup

* fix unit tests which used old syntax

* fix remaining unit tests

* replace __getattribute__ with getatt

* Address Cal's comment
Adjust docstring for the from_mbuild method.
Change docstring for the site.group.
Change MoleculeType and ResidueType to be NamedTuple.

* add options to infer (or not infer) the hierarchy structure when going from gmso to mbuild

* add infer_hierarchy for to_mbuild method

* parse group info when converting from mbuild

* WIP - removing all subtopology class and its reference

* remove remaining subtops from gmso objects and tests

* fix various errors/bugs, only 6 fails left

* fix parameterization bugs

* revert one step

* add patch for edge case, where molecule_tag is None

* fix case when site has no molecule or site

* trim misc codes

* make top.connections to be determined on the flight, remove self._connections

* Remove unnecessary function, relocate boundary bond assertion

* use n_direct_bonds inplace of is_independent when parsing residue

* add use_molecule_info option for apply

* add isomorphic check

* Modify Atomtyping parameterization to use flat molecule IDs for applying forcefields

* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci

* revert some changes

* Add match_ff_by, option to match by molecule name or group

* fix minor typo

* Fix bug, add remove_untyped
Fix bugs related to the new ff_match_by options.
Add new option to remove untyped connections.

* add missing flag

* more typos fixes

* add more unit tests for new features

* fix typo and add a comment

* parsing all lj and electrostatics scaling availabel, add unit tests

* change the error when molecule_id not in molecule_scaling_factors dict

* Add atom.clone and topology.create_subtop

* populate group and molecule info for cg atom

* fix typo

* turn error into warning when dict of ff is given on empty top

* remove return statement

* fix bug when apply ff with scaling factor of 0

* Mol2 format molecule information from RTI
This PR address concerns from @bc118 to automatically grab molecule
information from a mol2 file and attach it to the site.molecule for each
site in top.sites. This will go hand in hand with a future PR to address
the `gmso.formats.convert_mbuild.to_mbuild` utility in GMSO to properly
build an mBuild.Compound with the hierarchy that will results in
lossless mbuild to GMSO conversions.

* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci

* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci

* fix bug

* remove unused imports

Co-authored-by: Co Quach <daico007@gmail.com>
Co-authored-by: Co Quach <43968221+daico007@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

gmso/formats/mol2.py

@@ -6,6 +6,7 @@
 import unyt as u
 
 from gmso import Atom, Bond, Box, Topology
+from gmso.abc.abstract_site import MoleculeType, ResidueType
 from gmso.core.element import element_by_name, element_by_symbol
 from gmso.formats.formats_registry import loads_as
 
@@ -69,6 +70,7 @@ def from_mol2(filename, site_type="atom"):
         "@<TRIPOS>BOND": _parse_bond,
         "@<TRIPOS>CRYSIN": _parse_box,
         "@<TRIPOS>FF_PBC": _parse_box,
+        "@<TRIPOS>MOLECULE": _parse_molecule,
     }
     for section in sections:
         if section not in supported_rti:
@@ -79,7 +81,6 @@ def from_mol2(filename, site_type="atom"):
         else:
             supported_rti[section](topology, sections[section])
 
-    topology.update_topology()
     # TODO: read in parameters to correct attribute as well. This can be saved in various rti sections.
     return topology
 
@@ -139,13 +140,14 @@ def parse_ele(*symbols):
                     f"No charge was detected for site {content[1]} with index {content[0]}"
                 )
                 charge = None
-
+            molecule = top.label if top.__dict__.get("label") else top.name
             atom = Atom(
                 name=content[1],
                 position=position.to("nm"),
                 element=element,
                 charge=charge,
-                residue=(content[7], int(content[6])),
+                residue=ResidueType(content[7], int(content[6])),
+                molecule=MoleculeType(molecule, 1),
             )
             top.add_site(atom)
 
@@ -178,3 +180,8 @@ def _parse_box(top, section):
                 lengths=[float(x) for x in content[0:3]] * u.Å,
                 angles=[float(x) for x in content[3:6]] * u.degree,
             )
+
+
+def _parse_molecule(top, section):
+    """Parse molecule information from the mol2 file."""
+    top.label = str(section[0].strip())

gmso/tests/test_mol2.py

@@ -133,3 +133,28 @@ def test_neopentane_mol2_elements(self):
             r"consider manually adding the element to the topology$",
         ):
             top = Topology.load(get_fn("neopentane.mol2"))
+
+    def test_mol2_residues(self):
+        top = Topology.load(get_fn("parmed.mol2"))
+        assert np.all(
+            np.array([site.residue.name for site in top.sites]) == "RES"
+        )
+        assert np.all(
+            np.array([site.residue.number for site in top.sites]) == 1
+        )
+
+    def test_mol2_molecules(self):
+        top = Topology.load(get_fn("methane.mol2"))
+        assert np.all(
+            np.array([site.molecule.name for site in top.sites]) == "MET"
+        )
+        assert np.all(
+            np.array([site.molecule.number for site in top.sites]) == 1
+        )
+
+    def test_mol2_group(self):
+        # Is there a place to read from mol2 file?
+        top = Topology.load(get_fn("ethane.mol2"))
+        for site in top.sites:
+            site.group = "ethane"
+        assert np.all(np.array([site.group for site in top.sites]) == "ethane")