fix: nosubmit runs with default cluster config

.gitignore

@@ -69,5 +69,6 @@ Snakemake
 
 #temp test out
 examples/CyclophilinD_data/abfe/
-examples/abfe_out*
+examples/abfe_*out*
+examples/output
 /src/abfe/_version.py

environment_dev.yml

@@ -0,0 +1,34 @@
+name: abfe
+channels:
+  - bioconda
+  - openbiosim
+  - conda-forge
+  - defaults
+dependencies:
+  - python>=3.8
+  - pip
+  - conda-build
+
+  - snakemake=7.8.5
+  - tabulate=0.8.10
+
+  - gromacs>=2022.2
+
+  - matplotlib
+  - numpy=1.22.4
+  - scipy=1.7.3
+  - pandas
+
+  - mdanalysis
+  - parmed
+  - pdbfixer
+  - biosimspace
+  - openff-toolkit
+  - openff-interchange
+  - black
+
+  - pip:
+    - alchemlyb==2.0.0
+    - pymbar==4.0.1
+    - MDRestraintsGenerator
+

src/abfe/calculate_abfe.py

@@ -7,19 +7,33 @@
 from abfe.scripts import final_receptor_results
 
 
-def calculate_abfe(protein_pdb_path: str, ligand_sdf_paths: List[str], out_root_folder_path: str,
-                   approach_name: str = "", cofactor_sdf_path: str = None,
-                   n_cores_per_job: int = 8, num_jobs_receptor_workflow: int = None, num_jobs_per_ligand: int = 40, num_replicas: int = 3, small_mol_ff="openff",
-                   submit: bool = False, use_gpu: bool = True, hybrid_job: bool = True, cluster_config: dict = {}):
+def calculate_abfe(
+    protein_pdb_path: str,
+    ligand_sdf_paths: List[str],
+    out_root_folder_path: str,
+    approach_name: str = "",
+    cofactor_sdf_path: str = None,
+    n_cores_per_job: int = 8,
+    num_jobs_receptor_workflow: int = None,
+    num_jobs_per_ligand: int = 40,
+    num_replicas: int = 3,
+    small_mol_ff="openff",
+    submit: bool = False,
+    use_gpu: bool = True,
+    hybrid_job: bool = True,
+    cluster_config: dict = {},
+):
     orig_dir = os.getcwd()
     conf = {}
 
     # IO:
     ## Input standardization
     conf["input_protein_pdb_path"] = os.path.abspath(protein_pdb_path)
-    conf["input_ligands_sdf_path"] = [os.path.abspath(ligand_sdf_path) for ligand_sdf_path in ligand_sdf_paths]
+    conf["input_ligands_sdf_path"] = [
+        os.path.abspath(ligand_sdf_path) for ligand_sdf_path in ligand_sdf_paths
+    ]
 
-    if (cofactor_sdf_path is not None):
+    if cofactor_sdf_path is not None:
         conf["input_cofactor_sdf_path"] = os.path.abspath(cofactor_sdf_path)
     else:
         conf["input_cofactor_sdf_path"] = None
@@ -28,47 +42,63 @@ def calculate_abfe(protein_pdb_path: str, ligand_sdf_paths: List[str], out_root_
 
     ## Generate output folders
     for dir_path in [conf["out_approach_path"]]:
-        if (not os.path.isdir(dir_path)):
+        if not os.path.isdir(dir_path):
             os.mkdir(dir_path)
 
     # Prepare Input / Parametrize
     os.chdir(conf["out_approach_path"])
 
-    conf["ligand_names"] = [os.path.splitext(os.path.basename(sdf))[0] for sdf in conf["input_ligands_sdf_path"]]
-    conf["num_jobs"] = num_jobs_receptor_workflow if (num_jobs_receptor_workflow is not None) else len(conf["ligand_names"]) * num_replicas * 2
+    conf["ligand_names"] = [
+        os.path.splitext(os.path.basename(sdf))[0]
+        for sdf in conf["input_ligands_sdf_path"]
+    ]
+    conf["num_jobs"] = (
+        num_jobs_receptor_workflow
+        if (num_jobs_receptor_workflow is not None)
+        else len(conf["ligand_names"]) * num_replicas * 2
+    )
     conf["num_replica"] = num_replicas
-    conf['build_system'] = True
+    conf["build_system"] = True
     conf["small_mol_ff"] = small_mol_ff
 
     print("Prepare")
     print("\tstarting preparing ABFE-ligand file structur")
-    build_ligand_flows(input_ligand_paths=conf["input_ligands_sdf_path"],
-                       input_protein_path=conf["input_protein_pdb_path"],
-                       input_cofactor_path=conf["input_cofactor_sdf_path"],
-                       out_root_path=conf["out_approach_path"],
-                       num_max_thread=n_cores_per_job,
-                       num_replicas=num_replicas, num_jobs=num_jobs_per_ligand,
-                       cluster_config=cluster_config,
-                       use_gpu=use_gpu, hybrid_job=hybrid_job)
+    build_ligand_flows(
+        input_ligand_paths=conf["input_ligands_sdf_path"],
+        input_protein_path=conf["input_protein_pdb_path"],
+        input_cofactor_path=conf["input_cofactor_sdf_path"],
+        out_root_path=conf["out_approach_path"],
+        num_max_thread=n_cores_per_job,
+        num_replicas=num_replicas,
+        num_jobs=num_jobs_per_ligand,
+        cluster_config=cluster_config,
+        use_gpu=use_gpu,
+        hybrid_job=hybrid_job,
+    )
 
     print("\tstarting preparing ABFE-Approach file structure: ", out_root_folder_path)
     expected_out_paths = int(num_replicas) * len(conf["ligand_names"])
     result_paths = glob.glob(conf["out_approach_path"] + "/*/*/dG*tsv")
 
-    job_approach_file_path= None
-    if (len(result_paths) != expected_out_paths):
+    job_approach_file_path = None
+    if len(result_paths) != expected_out_paths:
         print("\tBuild approach struct")
-        job_approach_file_path = build_approach_flow(approach_name=approach_name,
-                                                     num_jobs=conf["num_jobs"],
-                                                     conf=conf, submit=submit,
-                                                     cluster_config=cluster_config)
+        job_approach_file_path = build_approach_flow(
+            approach_name=approach_name,
+            num_jobs=conf["num_jobs"],
+            conf=conf,
+            submit=submit,
+            cluster_config=cluster_config,
+        )
     print("Do")
     print("\tSubmit Job - ID: ", job_approach_file_path)
     # Final gathering
     print("\tAlready got results?: " + str(len(result_paths)))
-    if (len(result_paths) > 0):
+    if len(result_paths) > 0:
         print("Trying to gather ready results", out_root_folder_path)
-        final_receptor_results.get_final_results(out_dir=out_root_folder_path, in_root_dir=out_root_folder_path)
+        final_receptor_results.get_final_results(
+            out_dir=out_root_folder_path, in_root_dir=out_root_folder_path
+        )
 
     print()
     os.chdir(orig_dir)

src/abfe/calculate_abfe_gmx.py

@@ -10,76 +10,106 @@
 
 log = logging.getLogger(__file__)
 log.setLevel(logging.INFO)
-def calculate_abfe_gmx(input_dir:str, out_root_folder_path: str, approach_name: str = "",
-                   n_cores_per_job: int = 8, num_jobs_receptor_workflow: int = None, num_jobs_per_ligand: int = 40, num_replicas: int = 3,
-                   submit: bool = False, use_gpu: bool = True, hybrid_job: bool = True, cluster_config: dict = {}):
+
+
+def calculate_abfe_gmx(
+    input_dir: str,
+    out_root_folder_path: str,
+    approach_name: str = "",
+    n_cores_per_job: int = 8,
+    num_jobs_receptor_workflow: int = None,
+    num_jobs_per_ligand: int = 40,
+    num_replicas: int = 3,
+    submit: bool = False,
+    use_gpu: bool = True,
+    hybrid_job: bool = True,
+    cluster_config: dict = {},
+):
     orig_dir = os.getcwd()
     conf = {}
 
     # IO:
     ## Input standardization
     conf["input_path"] = os.path.abspath(input_dir)
-    conf['build_system'] = False
+    conf["build_system"] = False
     conf["out_approach_path"] = os.path.abspath(out_root_folder_path)
     conf["small_mol_ff"] = "custom"
 
     ## Generate output folders
     for dir_path in [conf["out_approach_path"]]:
-        if (not os.path.isdir(dir_path)):
+        if not os.path.isdir(dir_path):
             os.mkdir(dir_path)
 
     # Prepare Input / Parametrize
     os.chdir(conf["out_approach_path"])
 
-    #get Ligands:
-    conf["num_jobs"] = num_jobs_receptor_workflow if (num_jobs_receptor_workflow is not None) else len(os.listdir(conf["input_path"])) * num_replicas * 2
+    # get Ligands:
+    conf["num_jobs"] = (
+        num_jobs_receptor_workflow
+        if (num_jobs_receptor_workflow is not None)
+        else len(os.listdir(conf["input_path"])) * num_replicas * 2
+    )
     conf["num_replica"] = num_replicas
 
-    ligand_dirs = [conf["input_path"]+"/"+d for d in os.listdir(conf["input_path"]) if(os.path.isdir(conf["input_path"]+"/"+d))]
+    ligand_dirs = [
+        conf["input_path"] + "/" + d
+        for d in os.listdir(conf["input_path"])
+        if (os.path.isdir(conf["input_path"] + "/" + d))
+    ]
     print("Found ligands: ")
     log.info("Found ligands:")
     conf["ligand_names"] = []
     for ligand_dir in ligand_dirs:
         ligand_name = os.path.basename(ligand_dir)
         print("\t", ligand_name)
-        log.info("\t"+ligand_name)
+        log.info("\t" + ligand_name)
 
         new_lig_dir = conf["out_approach_path"] + "/" + ligand_name
 
-        if(not os.path.exists(new_lig_dir)):
+        if not os.path.exists(new_lig_dir):
             os.mkdir(new_lig_dir)
 
-        if(not os.path.exists(new_lig_dir+"/input")):
-            shutil.copytree(ligand_dir, new_lig_dir+"/input")
+        if not os.path.exists(new_lig_dir + "/input"):
+            shutil.copytree(ligand_dir, new_lig_dir + "/input")
 
         conf["ligand_names"].append(ligand_name)
-        build_replicas_simulation_flow(out_ligand_path=new_lig_dir,
-                                       input_ligand_path=new_lig_dir+"/input",
-                                       ligand_name=ligand_name,
-                                       num_max_thread=n_cores_per_job,
-                                       num_replicas=num_replicas, cluster_config=cluster_config, submit=False,
-                                       num_jobs=num_jobs_per_ligand,
-                                       use_gpu=use_gpu, hybrid_job=hybrid_job)
+        build_replicas_simulation_flow(
+            out_ligand_path=new_lig_dir,
+            input_ligand_path=new_lig_dir + "/input",
+            ligand_name=ligand_name,
+            num_max_thread=n_cores_per_job,
+            num_replicas=num_replicas,
+            cluster_config=cluster_config,
+            submit=False,
+            num_jobs=num_jobs_per_ligand,
+            use_gpu=use_gpu,
+            hybrid_job=hybrid_job,
+        )
 
     log.info("\tstarting preparing ABFE-Approach file structur: ", out_root_folder_path)
     expected_out_paths = int(num_replicas) * len(conf["ligand_names"])
     result_paths = glob.glob(conf["out_approach_path"] + "/*/*/dG*tsv")
 
-    if (len(result_paths) != expected_out_paths):
+    if len(result_paths) != expected_out_paths:
         print("\tBuild approach struct")
 
-        job_approach_file_path = build_approach_flow(approach_name=approach_name,
-                                                     num_jobs=conf["num_jobs"],
-                                                     conf=conf, submit=submit,
-                                                     cluster_config=cluster_config)
+        job_approach_file_path = build_approach_flow(
+            approach_name=approach_name,
+            num_jobs=conf["num_jobs"],
+            conf=conf,
+            submit=submit,
+            cluster_config=cluster_config,
+        )
         print("Do")
         print("\tSubmit Job - ID: ", job_approach_file_path)
 
     # Final gathering
     print("\tAlready got results?: " + str(len(result_paths)))
-    if (len(result_paths) > 0):
+    if len(result_paths) > 0:
         print("Trying to gather ready results", out_root_folder_path)
-        final_receptor_results.get_final_results(out_dir=out_root_folder_path, in_root_dir=out_root_folder_path)
+        final_receptor_results.get_final_results(
+            out_dir=out_root_folder_path, in_root_dir=out_root_folder_path
+        )
 
     print()
     os.chdir(orig_dir)

src/abfe/conf.py

@@ -7,12 +7,11 @@
 
 
 std_conf = {
-    #Gromacs Kernels
+    # Gromacs Kernels
     "gmx_kernel_cpu": template.gmx_submit_kernels_path + "/def_cpu_job.sh",
     "gmx_kernel_cpu_cont": template.gmx_submit_kernels_path + "/def_cpu_job_cont.sh",
     "gmx_kernel_gpu": template.gmx_submit_kernels_path + "/def_gpu_job.sh",
     "gmx_kernel_gpu_cont": template.gmx_submit_kernels_path + "/def_gpu_job_cont.sh",
-
-    #GMX flag addition
-    "gmx_add_flag":"",
-            }
+    # GMX flag addition
+    "gmx_add_flag": "",
+}