Merge pull request #496 from carlocamilloni/main

small fixes, more regtests
multi-ego · Dec 5, 2024 · f3efb0c · f3efb0c
2 parents 4dab8d7 + 86cfab5
commit f3efb0c
Show file tree

Hide file tree

Showing 44 changed files with 9,805 additions and 20 deletions.
diff --git a/multi-ego-ready.ff/README.md b/multi-ego-ready.ff/README.md
diff --git a/src/multiego/ensemble.py b/src/multiego/ensemble.py
@@ -526,7 +526,7 @@ def generate_14_data(meGO_ensemble):
     # First of all we generate the random-coil 1-4 interactions:
     pairs14 = pd.DataFrame()
     exclusion_bonds14 = pd.DataFrame()
-    for molecule, bond_pair in meGO_ensemble["bond_pairs"].items():
+    for idx, (molecule, bond_pair) in enumerate(meGO_ensemble["bond_pairs"].items(), start=1):
         if not bond_pair:
             continue
         reduced_topology = (
@@ -604,6 +604,10 @@ def generate_14_data(meGO_ensemble):
             tmp["ai"], tmp["aj"] = tmp["aj"], tmp["ai"]
             pairs = pd.concat([pairs, tmp], axis=0, sort=False, ignore_index=True)
 
+        mol_ai = f"{idx}_{molecule}"
+        pairs["molecule_name_ai"] = mol_ai
+        pairs["molecule_name_aj"] = mol_ai
+
         if not pairs.empty:
             pairs14 = pd.concat([pairs14, pairs], axis=0, sort=False, ignore_index=True)
 
@@ -1220,6 +1224,9 @@ def generate_LJ(meGO_ensemble, train_dataset, basic_LJ, parameters):
     # rename the columns _x
     meGO_LJ_14.columns = meGO_LJ_14.columns.str.rstrip("_x")
 
+    # remove intermolecular interactions across molecules from meGO_LJ_14
+    meGO_LJ_14 = meGO_LJ_14[meGO_LJ_14["molecule_name_ai"] == meGO_LJ_14["molecule_name_aj"]]
+
     # copy 1-4 interactions into meGO_LJ_14
     copy14 = meGO_LJ.loc[(meGO_LJ["1-4"] == "1_4")]
     meGO_LJ_14 = pd.concat([meGO_LJ_14, copy14], axis=0, sort=False, ignore_index=True)
@@ -1367,7 +1374,7 @@ def make_pairs_exclusion_topology(meGO_ensemble, meGO_LJ_14):
         Contains the "write out"-ready pairs-exclusions interactions for each molecule
     """
     pairs_molecule_dict = {}
-    for molecule, bond_pair in meGO_ensemble["bond_pairs"].items():
+    for idx, (molecule, bond_pair) in enumerate(meGO_ensemble["bond_pairs"].items(), start=1):
         reduced_topology = (
             meGO_ensemble["topology_dataframe"]
             .loc[meGO_ensemble["topology_dataframe"]["molecule_name"] == molecule][
@@ -1394,8 +1401,9 @@ def make_pairs_exclusion_topology(meGO_ensemble, meGO_LJ_14):
         exclusion_bonds, p14 = topology.get_14_interaction_list(reduced_topology, bond_pair)
         pairs = pd.DataFrame()
         if not meGO_LJ_14.empty:
+            mol_ai = f"{idx}_{molecule}"
             # pairs do not have duplicates because these have been cleaned before
-            pairs = meGO_LJ_14[
+            pairs = meGO_LJ_14[meGO_LJ_14["molecule_name_ai"] == mol_ai][
                 [
                     "ai",
                     "aj",

diff --git a/test/run_make_mat.sh b/test/run_make_mat.sh
@@ -12,3 +12,11 @@ tar -zxf test_inputs/make_mat_popc/hh.tgz -C test_inputs/make_mat_popc/
 python ../tools/make_mat/make_mat.py --histo test_inputs/make_mat_popc/histo --target_top test_inputs/make_mat_popc/topol_md.top --mego_top test_inputs/make_mat_popc/topol_ref.top --cutoff 0.75 --mode intra --out  test_inputs/make_mat_popc/ 
 diff <(gzip -dc test_inputs/make_mat_popc/intramat_1_1.ndx.gz) <(gzip -dc test_outputs/make_mat_popc/intramat_1_1.ndx.gz)
 
+tar -zxf test_inputs/make_mat_het_trim/hh.tgz -C test_inputs/make_mat_het_trim/
+python ../tools/make_mat/make_mat.py --histo test_inputs/make_mat_het_trim/histo --target_top test_inputs/make_mat_het_trim/topol_mego.top --mego_top test_inputs/make_mat_het_trim/topol.top --cutoff 0.75 --out  test_inputs/make_mat_het_trim/ 
+diff <(gzip -dc test_inputs/make_mat_het_trim/intramat_1_1.ndx.gz) <(gzip -dc test_outputs/make_mat_het_trim/intramat_1_1.ndx.gz)
+diff <(gzip -dc test_inputs/make_mat_het_trim/intramat_2_2.ndx.gz) <(gzip -dc test_outputs/make_mat_het_trim/intramat_2_2.ndx.gz)
+diff <(gzip -dc test_inputs/make_mat_het_trim/intramat_3_3.ndx.gz) <(gzip -dc test_outputs/make_mat_het_trim/intramat_3_3.ndx.gz)
+diff <(gzip -dc test_inputs/make_mat_het_trim/intermat_1_2.ndx.gz) <(gzip -dc test_outputs/make_mat_het_trim/intermat_1_2.ndx.gz)
+diff <(gzip -dc test_inputs/make_mat_het_trim/intermat_1_3.ndx.gz) <(gzip -dc test_outputs/make_mat_het_trim/intermat_1_3.ndx.gz)
+diff <(gzip -dc test_inputs/make_mat_het_trim/intermat_2_3.ndx.gz) <(gzip -dc test_outputs/make_mat_het_trim/intermat_2_3.ndx.gz)
diff --git a/test/test_cases.txt b/test/test_cases.txt
@@ -4,3 +4,5 @@
 --config TEST_ROOT/test_inputs/ttrref/config.yml --explicit_name case # --system ttrref --egos production
 --config TEST_ROOT/test_inputs/lyso-bnz_ref/config_1.yml --explicit_name case # --system lyso-bnz_ref --egos production
 --config TEST_ROOT/test_inputs/lyso-bnz_ref/config_2.yml --explicit_name case # --system lyso-bnz_ref --egos production
+--system het_trim --egos rc --no_header --explicit_name case
+--system het_trim --egos production --epsilon 0.3 --train train --no_header --explicit_name case