1. Adding QDESCP-xTB and QCORR-ORCA pytests

Author: jvalegre

Example_workflows/QCORR_processing_QM_outputs/QCORR_7/orca_TS_success.out

@@ -1,13 +1,15 @@
 {
  "cells": [
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "#### Organic molecule with SMILES input, CSEARCH performs conformational sampling with RDKit, QDESCP produces xtb properties\n"
    ]
   },
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
@@ -35,6 +37,7 @@
    ]
   },
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
@@ -54,7 +57,7 @@
     "# 1) Program xTB (program='xtb')\n",
     "# 1) Files to run (files=sdf_rdkit_files)\n",
     "# 2) Include Boltzmann averaging and RDKit descriptors (boltz=True)\n",
-    "qdescp(program='xtb',files=sdf_rdkit_files,boltz=True)\n",
+    "qdescp(program='xtb',files=sdf_rdkit_files)\n",
     " "
    ]
   },

Example_workflows/QCORR_processing_QM_outputs/QCORR_7/orca_imag_freq.out

@@ -216,6 +216,7 @@ def main():
     if args.qdescp:
         qdescp(
             verbose=args.verbose,
+            command_line=args.command_line,
             w_dir_main=args.w_dir_main,
             destination=args.destination,
             files=args.files,

Example_workflows/QDESCP_generating_xtb_features/QDESCP_features.ipynb

@@ -101,9 +101,9 @@
 import time
 from aqme.utils import (
     load_variables,
-    substituted_mol,
     mol_from_sdf_or_mol_or_mol2,
-    add_prefix_suffix
+    add_prefix_suffix,
+    check_xtb
 )
 from aqme.filter import ewin_filter, pre_E_filter, RMSD_and_E_filter
 from aqme.cmin_utils import creation_of_dup_csv_cmin
@@ -140,6 +140,10 @@ def __init__(self, **kwargs):
             self.args.log.finalize()
             sys.exit()
 
+        # check if xTB is installed
+        if self.args.program.lower() == "xtb":
+            _ = check_xtb(self)
+
         # retrieves the different files to run in CMIN
         if len(self.args.files) == 0:
             self.args.log.write('\nx  No files were found! Make sure you use quotation marks if you are using * (i.e. --files "*.sdf")')
@@ -421,10 +425,10 @@ def compute_cmin(self, file):
 
         return dup_data
 
-    # xTB AND ANI MAIN OPTIMIZATION PROCESS
+    # ANI MAIN OPTIMIZATION PROCESS
     def ani_optimize(self, mol, charge, mult):
 
-        # Attempts ANI/xTB imports and exits if the programs are not installed
+        # Attempts ANI imports and exits if the programs are not installed
         try:
             import torch
             import warnings
@@ -509,27 +513,14 @@ def get_cmin_model(self):
         Function to generate the optimization model for CMIN (using xTB or ANI methods)
         """
 
-        if self.args.program.lower() == "ani":
-            try:
-                import torchani
-            except (ImportError,ModuleNotFoundError):
-                self.args.log.write("x  Torchani is not installed! You can install the program with 'pip install torchani'")
-                self.args.log.finalize()
-                sys.exit()
-
-            model = getattr(torchani.models,self.args.ani_method)()
+        try:
+            import torchani
+        except (ImportError,ModuleNotFoundError):
+            self.args.log.write("x  Torchani is not installed! You can install the program with 'pip install torchani'")
+            self.args.log.finalize()
+            sys.exit()
 
-        elif self.args.program.lower() == "xtb":
-            try:
-                subprocess.run(
-                    ["xtb", "-h"], stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL
-                )
-            except FileNotFoundError:
-                self.args.log.write("x  xTB is not installed (CREST cannot be used)! You can install the program with 'conda install -c conda-forge xtb'")
-                self.args.log.finalize()
-                sys.exit()
-    
-            model = None
+        model = getattr(torchani.models,self.args.ani_method)()
 
         return model
 

aqme/aqme.py

@@ -212,6 +212,7 @@
     load_variables,
     set_metal_atomic_number,
     check_xtb,
+    check_crest,
     get_files
     )
 from aqme.csearch.crest import xtb_opt_main
@@ -246,6 +247,7 @@ def __init__(self, **kwargs):
 
         if self.args.program.lower() == "crest":
             _ = check_xtb(self)
+            _ = check_crest(self)
 
         if self.args.smi is None and self.args.input == "":
             self.args.log.write("\nx  Program requires either a SMILES or an input file to proceed! Please look up acceptable file formats. Specify: smi='CCC' (or input='filename.csv')")
@@ -351,7 +353,12 @@ def load_jobs(self, csearch_file):
 
         # Prepare the jobs
         prepare_function = Extension2inputgen[file_format]
-        job_inputs = prepare_function(self.args, csearch_file)
+        try:
+            job_inputs = prepare_function(self.args, csearch_file)
+        except FileNotFoundError:
+            self.args.log.write(f'\nx  File {os.path.basename(csearch_file)} was not found! In the "input" option, make sure that 1) the PATH to the files is correct and 2) the PATH doesn\'t start with "/".')
+            self.args.log.finalize()
+            sys.exit()     
 
         return job_inputs
 

aqme/cmin.py

@@ -239,7 +239,7 @@ def xtb_opt_main(
             run_command(command, f"{xyzin.split('.')[0]}_xtb1.out")
             os.rename(str(dat_dir) + "/xtbopt.xyz", xyzoutxtb1)
         except FileNotFoundError:
-            self.args.log.write(f"\nx   There was an error during the xTB pre-optimization. This error is related to parallelization of xTB jobs and is normally observed when using metal complexes in some operative systems/OpenMP versions. AQME is switching to using one processor (nprocs=1).\n")
+            self.args.log.write(f"\nx  There was an error during the xTB pre-optimization. This error is related to parallelization of xTB jobs and is normally observed when using metal complexes in some operative systems/OpenMP versions. AQME is switching to using one processor (nprocs=1).\n")
             self.args.nprocs = 1
             try:
                 if self.args.xtb_keywords is None:
@@ -252,9 +252,9 @@ def xtb_opt_main(
                 os.rename(str(dat_dir) + "/xtbopt.xyz", xyzoutxtb1)
             except FileNotFoundError:
                 if self.args.program.lower() == "crest":
-                    self.args.log.write(f"\nx   There was another error during the xTB pre-optimization that could not be fixed. Trying CREST directly with no xTB preoptimization.\n")
+                    self.args.log.write(f"\nx  There was another error during the xTB pre-optimization that could not be fixed. Trying CREST directly with no xTB preoptimization.\n")
                 else:
-                    self.args.log.write(f"\nx   There was another error during the xTB pre-optimization that could not be fixed (this molecule will be skipped).\n")
+                    self.args.log.write(f"\nx  There was another error during the xTB pre-optimization that could not be fixed (this molecule will be skipped).\n")
                 cmin_valid = False
                 mol_rd = None
 
@@ -324,7 +324,7 @@ def xtb_opt_main(
             else:
                 shutil.copy(str(dat_dir) + "/crest_conformers.xyz", xyzoutall)
         except FileNotFoundError:
-            self.args.log.write("\nx   CREST conformer sampling failed! Please, try other options (i.e. include constrains, change the crest_keywords option, etc.)")
+            self.args.log.write("\nx  CREST conformer sampling failed! Please, try other options (i.e. include constrains, change the crest_keywords option, etc.)")
             cmin_valid = False
 
     if cmin_valid:

aqme/csearch/base.py

@@ -325,15 +325,15 @@ def filters(mol, log, molwt_cutoff):
     # Filter 1
     molwt_cutoff = float(molwt_cutoff)
     if Descriptors.MolWt(mol) >= molwt_cutoff and molwt_cutoff > 0:
-        log.write(f"x   Skipping this molecule as total molar mass > {molwt_cutoff}")
+        log.write(f"x  Skipping this molecule as total molar mass > {molwt_cutoff}")
         return False
 
     # Filter 2
     symbols = [atom.GetSymbol() for atom in mol.GetAtoms()]
 
     unknown_atoms = list(set(symbols) - set(periodic_table()))
     if unknown_atoms:
-        log.write(f" Exiting as atoms [{','.join(unknown_atoms)}] are not in the periodic table")
+        log.write(f"x  Exiting as atoms [{','.join(unknown_atoms)}] are not in the periodic table")
         return False
 
     # Passed

aqme/csearch/crest.py

@@ -617,7 +617,7 @@ def get_json_data(self, file, cclib_data):
                 cclib_data["metadata"]["keywords line"] = keywords_line[1:].rstrip("\n")
                 calc_type = "ground_state"
                 for keyword in keywords_line.split():
-                    if keyword.lower() in ["OptTS",'NEB-TS']:
+                    if keyword.lower() in ["optts",'neb-ts']:
                         calc_type = "transition_state"
                         break
                     if keyword.lower()[0:3] == 'pal':